Finnish Food Safety Authority, Evira Research Department

Microbiology Unit Helsinki, Finland

and

Department of Food Hygiene and Environmental Health Faculty of Veterinary Medicine

University of Helsinki Helsinki, Finland

Finnish cattle as reservoir of Campylobacter spp.

Marjaana Hakkinen

Academic Dissertation

To be presented with the permission of the Faculty of Veterinary Medicine, University of Helsinki, for public examination in Auditorium XII, Fabianinkatu 33, Helsinki

on 26 November 2010, at 12 o’clock noon HELSINKI 2010

Supervising professor Marja-Liisa Hänninen, DVM, PhD, Professor Department of Food Hygiene and Environmental Health Faculty of Veterinary Medicine University of Helsinki Helsinki, Finland Supervised by Marja-Liisa Hänninen, DVM, PhD, Professor Department of Food Hygiene and Environmental Health Faculty of Veterinary Medicine University of Helsinki Helsinki, Finland Reviewed by Thomas Alter, PhD, Professor Department of Veterinary Medicine Institute of Food Hygiene Freie Universität Berlin Berlin, Germany Eva Møller Nielsen, PhD Department of Microbiological Surveillance and Disease Statens Serum Institut Copenhagen, Denmark Opponent Wilma Jacobs-Reitsma, PhD National Institute for Public Health and Environment The Netherlands ISSN 1796-4660, ISBN 978-952-225-076-6 (print) ISSN 1797-2981, ISBN 978-952-225-077-3 (pdf) Helsinki University Printing House 2010

Abstract

The reported incidence of human campylobacteriosis in Finland is higher than in most other European countries. A high annual percentage of sporadic infections is of foreign origin, although a notable proportion of summer infections is domestically acquired. While chickens appear to be a major source of campylobacters for humans in most countries, the prevalence of campylobacters is very low in chicken slaughter batches in Finland. Data on other potential animal reservoirs of human pathogenic campylobacters in Finland are scarce. Consequently, this study aimed to investigate the status of Finnish cattle as a potential source of thermophilic Campylobacter spp. and antibiotic-resistant Campylobacter jejuni for human sporadic campylobacter infections of domestic origin. A survey of the prevalence of thermophilic Campylobacter spp. in Finnish cattle studied bovine rectal faecal samples (n=952) and carcass surface samples (n=948) from twelve Finnish slaughterhouses from January to December 2003. The total prevalence of Campylobacter spp. in faecal samples was 31.1%, and in carcass samples 3.5%. Campylobacter jejuni, the most common species, was present in 19.5% of faecal samples and in 3.1% of carcasses. In addition to thermophilic Campylobacter spp., C. hyointestinalis ssp. hyointestinalis was present in bovine samples. The prevalence of campylobacters was higher among beef cattle than among dairy cattle. Using the enrichment method, the number of positive faecal samples was 7.5 times higher than that obtained by direct plating. The predominant serotypes of faecal C. jejuni, determined by serotyping with a set of 25 commercial antisera for heat-stable antigens (Penner), were Pen2 and Pen4-complex, which covered 52% of the samples. Genotyping with pulsed-field gel electrophoresis (PFGE) using SmaI restriction yielded a high diversity of C. jejuni subtypes in cattle. Determining the minimum inhibitory concentrations of ampicillin, enrofloxacin, erythromycin, gentamicin, nalidixic acid, and oxytetracycline among bovine C. jejuni isolates using a commercial broth microdilution method yielded 9% of isolates resistant to at least one of the antimicrobials examined. No multiresistant isolates were found among the bovine C. jejuni strains. The study of the shedding patterns of Campylobacter spp. among three Finnish dairy cattle herds included the examination of fresh faecal samples and tank milk samples taken five times, as well as samples from drinking troughs taken once during the one-year study. The semiquantitative enrichment method detected C. jejuni in 169 of the 340 faecal samples, mostly at low levels. In addition, C. jejuni was present in one drinking trough sample. The prevalence between herds and sampling occasions varied widely. PFGE, using SmaI as restriction enzyme, identified only a few subtypes in each herd. In two

2

of the herds, two subtypes persisted throughout the sampling. Individual animals presented various shedding patterns during the study. Comparison of C. jejuni isolates from humans, chickens and cattle included the design of primers for four new genetic markers selected from completely sequenced C. jejuni genomes 81-176, RM1221 and NCTC 11168, and the PCR examination of domestic human isolates from southern Finland in 1996, 2002 and 2003 (n=309), chicken isolates from 2003, 2006 and 2007 (n=205), and bovine isolates from 2003 (n=131). The results revealed that bovine isolates differed significantly from human and chicken isolates. In particular, the �-glutamyl transpeptidase gene was uncommon among bovine isolates. The PFGE genotyping of C. jejuni isolates, using SmaI and KpnI restriction enzymes, included a geographically representative collection of isolates from domestic sporadic human infections, chicken slaughter batches, and cattle faeces and carcasses during the seasonal peak of campylobacteriosis in the summer of 2003. The study determined that 55.4% of human isolates were indistinguishable from those of chickens and cattle. Temporal association between isolates from humans and chickens was possible in 31.4% of human infections. Approximately 19% of the human infections may have been associated with cattle. However, isolates from bovine carcasses and human cases represented different PFGE subtypes. In conclusion, this study suggests that Finnish cattle is a notable reservoir of C. jejuni, the most important Campylobacter sp. in human enteric infections. Although the concentration of these organisms in bovine faeces appeared to be low, excretion can be persistent. The genetic diversity and presence or absence of marker genes support previous suggestions of host-adapted C. jejuni strains, and may indicate variations in virulence between strains from different hosts. In addition to chickens, Finnish cattle appeared to be an important reservoir and possible source of C. jejuni in domestic sporadic human infections. However, sources of campylobacters may differ between rural and urban areas in Finland, and in general, the transmission of C. jejuni of bovine origin probably occurs via other routes than food.

3

Acknowledgements

This study begun at the National Veterinary and Food Research Institute (EELA) and ended at the Microbiology unit of Finnish Food Safety Authority (Evira). I thank Professor Tuula Honkanen-Buzalski, previously the general director of EELA and the current head of the Research Department in Evira, and DVM PhD Vesa Myllys, the former head of the Microbiology Unit, for the opportunity to carry out this study. The financial support from the Finnish Veterinary Science Foundation, the Walter Ehrström Foundation and the University of Helsinki made this work possible. I thank these organisations for their funding, which enabled me to take study leave to write the articles and the thesis. My warmest appreciation goes to my supervisor Professor Marja-Liisa Hänninen for her continuous encouragement and excellent scientific guidance during these years. I truly admire her vast knowledge and enthusiasm in the field of science. Professor Thomas Alter and PhD Eva Møller Nielsen, the official reviewers of my thesis, deserve my sincere thanks for their constructive and valuable criticism, as does Stephen Stalter for his excellent revision of the English language of my thesis. I am also grateful to my co-authors Helmi Heiska, Manuel Gonzalez, Ulla-Maija Nakari, and Professors Hilpi Rautelin and Anja Siitonen for their valuable contributions to this project and for our fruitful co-operation. I warmly thank the veterinary inspectors for collecting the samples at the slaughterhouses, and the farmers for permitting collect samples from their barns. Without their contribution this study would have been impossible. Completion of the project depended on the excellent work of the laboratory staff. I am especially grateful to Kirsi Eklund for her laboratory work throughout the project. Mira Kankare, Lea Nygård, Kaija Pajunen, Maaret Hyppönen, Mirva Tahvanainen, Sari Maljanen, Katriina Mälkönen, Marja Rasinperä, and all the young students who spent their summer hunting campylobacters in our laboratory also deserve my warm thanks for their skilful assistance at different stages of this project. I offer my gratitude to all of my colleagues at the Microbiology Unit for their support and encouragement. My special thanks belong to Leila Rantala for introducing me to the secrets of Bionumerics, to Tuula Johansson and Saija Hallanvuo for handling my share while I was absent, and to Ansu Myllyniemi for her friendship and tireless optimism.

4

I am also grateful to all of my friends for their friendship and for reminding me of other important aspects of life during these years so full of work. My very special thanks go to my aunt Maija-Leena, whose enquiring character I wish to emulate. My late parents always encouraged me to study and supported me in many ways, and my late sister left me an unforgettable memory of sisterhood: I miss her so much. I am deeply grateful to my parents, my sister and my brother. My deepest gratitude goes to my dearest and nearest: my husband Jukka and our daughters Elisa and Tuuli. With you I have experienced the happiest times in my life.

5

Contents

Abstract 1 Acknowledgements 3 Contents 5 List of original publications 7 Abbreviations 8

1 Introduction 9 2 Review of the literature 11 2.1 Campylobacter spp. and human enteric diseases 11 2.1.1 Human campylobacteriosis 11 2.1.2 Sources of Campylobacter spp. for human infection 12 2.2 Subtyping of Campylobacter jejuni 14 2.2.1 Serotyping 14 2.2.2 Pulsed-field gel electrophoresis (PFGE) 15 2.2.3 Amplified fragment length polymorphism (AFLP) 16 2.2.4 Fla-SVR typing 16 2.2.5 Multilocus sequence typing (MLST) 16 2.2.6 DNA microarray 17 2.3 Campylobacter spp. in cattle 18 2.3.1 Prevalence of Campylobacter spp. in cattle 18 2.3.2 Campylobacter species in cattle 20 2.3.3 Campylobacter jejuni in cattle at farm 20 2.3.4 Genetic diversity and host adaptation of bovine

Campylobacter jejuni strains 22 2.4 Campylobacter spp. in foods of bovine origin 22 2.4.1 Beef and edible offal 22 2.4.2 Milk and milk products 24 2.5 Cattle as a source of Campylobacter spp. in human infections 25 2.5.1 Outbreak investigations and case-control studies 25 2.5.2 Genotyping and source attribution studies 26 2.6 Antimicrobial resistance of Campylobacter spp. 26 3 Aims of the study 29 4 Materials and methods 30 4.1 Sampling 30 4.2 Isolation of Campylobacter spp. (I, II) 30 4.3 Campylobacter jejuni isolates (III, IV) 31 4.3.1 Human isolates (III, IV) 31 4.3.2 Chicken isolates (III, IV) 32 4.3.3 Bovine isolates (III, IV) 32 4.4 Serotyping of Campylobacter jejuni isolates (I) 32 4.5 Pulsed-field gel electrophoresis (I, II, IV) 34 4.6 PCR of genetic markers of Campylobacter jejuni (III) 34 4.7 Determination of antimicrobial susceptibility of

Campylobacter jejuni isolates (I) 35

6

4.8 Statistical methods 35

5 Results 36 5.1 Prevalence of Campylobacter spp. in cattle at slaughter and on

three dairy farms (I, II) 36 5.1.2 Subtypes of Campylobacter jejuni (I, II, IV) 39 5.1.3 Occurrence of genetic markers among Campylobacter jejuni

isolates from humans, chickens and cattle (III) 43 5.1.4 Antimicrobial susceptibility of bovine Campylobacter

jejuni isolates (I) 43

6 Discussion 45 6.1 Campylobacter spp. in Finnish cattle 45 6.2 The diversity of Campylobacter jejuni in Finnish cattle 46 6.3 Chickens and cattle as sources of Campylobacter jejuni

in sporadic human infections in Finland 47 6.3.1 Comparison of subtypes of Campylobacter jejuni from human

infections, chickens and cattle 47 6.3.2 Genetic markers in differentiation of the sources of

Campylobacter jejuni in human infections 49 6.4 Antimicrobial susceptibility of bovine Campylobacter jejuni

isolates 50

7 Conclusions 51 References 53

7

List of original publications

This thesis is based on the following publications: I Hakkinen, M., Heiska, H. & Hänninen M.-L. 2007.

Prevalence of Campylobacter spp. in cattle in Finland and antimicrobial susceptibilities of bovine Campylobacterjejuni strains. Appl. Environ. Microbiol. 73, 3232-3238.

II Hakkinen, M. & Hänninen, M.-L. 2009. Shedding of

Campylobacter spp. in Finnish cattle on dairy farms. J. Appl. Microbiol. 107, 898-905.

III Gonzalez, M., Hakkinen, M., Rautelin, H. & Hänninen,

M.-L. 2009. Bovine Campylobacter jejuni strains differ from human and chicken strains in an analysis of certain molecular genetic markers, Appl. Environ. Microbiol. 75, 1208-10.

IV Hakkinen, M., Nakari U.-M. & Siitonen, A. 2009.

Chickens and cattle as sources of sporadic, domestically acquired Campylobacter jejuni infections in Finland. Appl. Environ. Microbiol. 75, 5244-5249.

The publications are indicated in the text by their roman numerals. The original articles have been reprinted with the permission of their copyright holders: The American Society for Microbiology (I, III and IV) and John Wiley and Sons (II).

8

Abbreviations

AFLP amplified fragment length polymorphism ATCC American Type Culture Collection BIOHAZ EFSA Panel on Biological Hazards bp base pair CDC Centers for Disease Control and Prevention CI confidence interval CLSI Clinical and Laboratory Standards Institute DMSO dimethyl sulfoxide oxidoreductase DNA deoxyribonucleic acid EDTA ethylenediamine tetraacetic acid EFSA European Food Safety Authority EUCAST European Committee for Antimicrobial Susceptibility

Testing flaA flagellin A gene ggt �-glutamyl transpeptidase gene ISO International Organization for Standardization mCCDA modified charcoal cefoperazone deoxycholate agar MIC minimum inhibitory concentration MLST multilocus sequence typing MPN most probable number NCFA Nordic Committee on Food Analysis NCTC National Collection of Type Cultures PCR polymerase chain reaction PFGE pulsed-field gel electrophoresis ST sequence type SVR short variable region THL National Institute for Health and Welfare TSI triple-sugar iron UV ultra violet WHO World Health Organization

9

1 Introduction

The genus Campylobacter was established in 1963 (Sebald and Véron 1963). Over a hundred years ago, however, scientists had already described these Vibrio-like organisms, which were present primarily in bovine and ovine abortions (Smith and Taylor 1919, Skirrow 2006), and occasionally in human disease as well (Levy 1946, King 1957). The importance of thermophilic campylobacters, and of Campylobacter jejuni and C. coli in particular, as human enteric pathogens has become clear since the 1970s, after the discovery of selective isolation methods for these fastidious organisms (Dekeyser et al. 1972, Butzler et al. 1973, Skirrow 1977). Subsequent intensive research has revealed that C. jejuni is the most common cause of human bacterial gastroenteritis worldwide (Baker et al. 2007, EFSA 2010b). Campylobacter infection, campylobacteriosis, is usually a self-limiting disease with clinical symptoms similar to those of other acute bacterial enteric infections (Blaser and Engberg 2008). The infective dose can be low: in experimental infections, a dose of 500 bacterial cells was sufficient to cause disease (Black et al. 1988), and outbreak data modelling has suggested that even fewer than 20 cells can induce symptoms (Teunis et al. 2005). Musculosceletal symptoms are common complications in connection with C. jejuni enteric infections, and reactive arthritis occurs in about 4% to 5% of cases (Doorduyn et al. 2008, Schönberg-Norio et al. 2009). The most serious, though infrequent sequela is Guillain-Barré syndrome, an acute neuromuscular paralysis (Jacobs et al. 2008). Campylobacters have a wide range of animal hosts, including food production animals, which can be carriers of these bacteria in their intestinal tract without showing clinical symptoms (Nielsen et al. 1997, Stanley and Jones 2003, Brown et al. 2004, Devane et al. 2005, Milnes et al. 2008). A particularly favourable environment for the proliferation of C. jejuni is the avian intestine (Lee and Newell 2006), and accordingly, poultry appears to be a major source of C. jejuni in humans (EFSA Panel on Biological Hazards [BIOHAZ] 2010). The prevalence of campylobacters in Finnish chicken slaughter batches is among the lowest in Europe (EFSA 2010c). The incidence of human campylobacteriosis in Finland, however, is among the highest in Europe, although the incidences may not be fully comparable between countries due to differences in their reporting systems (EFSA 2010b). Reporting is comparable between Nordic countries, however; the highest incidence of human campylobacter infections occurred in Finland (84/100 000 population), but was substantially lower in Norway (60.7/100 000 population), whereas the prevalence of campylobacters in chicken slaughter batches was low in

10

Finland (3.9%) and the lowest in Norway (3.2%) (EFSA 2010b, EFSA 2010c). A high percentage, up to 77% in 2008 ((National Institute for Health and Welfare 2009), of human campylobacter infections reported in Finland originate from travel abroad. Nevertheless, the proportion of domestically acquired campylobacter infections peaks in the summer season, comprising approximately 40% to 70% of reported cases (Vierikko et al. 2004, National Public Health Institute 2005). Similarly, the prevalence of campylobacters in Finnish chicken slaughter batches also peaks in late summer, whereas campylobacters are rarely found in chickens in winter. The prevalence of Campylobacter spp. in chicken slaughter batches has remained low with no major changes after the implementation of the Finnish campylobacter monitoring programme for chickens in June 2004 (http://www.zoonoosikeskus.fi/attachments/zoonoosit/kampylobakteeri/kampylobakteeri_2.pdf). Between 2000 and 2008, only three of the ten food-related campylobacteriosis outbreaks identified in Finland were attributed to chicken, turkey or duck meat (Finnish Food Safety Authority, unpublished data). On the other hand, the sources of sporadic campylobacter infections, which constitute the majority of cases, usually remain unclear. According to a recent Finnish study (Schönberg-Norio et al. 2006), the sources of domestically acquired campylobacteriosis differ depending on the age of the patient and the geographical area. But besides chickens (Hänninen et al. 2000, Perko-Mäkelä et al. 2002, Kärenlampi et al. 2003), available data on other potential animal reservoirs of campylobacters in Finland are limited.

11

2 Review of the literature

2.1 Campylobacter spp. and human enteric diseases

2.1.1 Human campylobacteriosis

Campylobacter spp., especially Campylobacter jejuni, is the most frequently reported cause of human bacterial gastroenteric infections. The incidence of campylobacteriosis has been steadily rising in most countries where the disease is notifiable (Baker et al. 2007, EFSA 2010b). Reports from New Zealand have presented the highest incidences between 2000 and 2007, peaking at 422.4 cases per 100 000 people in 2006 (Baker et al. 2006, Baker et al. 2007, Mullner et al. 2010a). The EFSA report on zoonoses in 2008 (EFSA 2010b) reported incidences of campylobacteriosis from <0.1 to 193.3/100 000 population in European countries. The wide variation among countries likely reflects differences in health care and reporting systems, and in microbiological methods rather than real differences in the incidence of campylobacter infections (Olson et al. 2008, Vally et al. 2009, EFSA 2010b). The majority of human cases are sporadic or small-scale family outbreaks, whereas large outbreaks occur infrequently (Olson et al. 2008). Identification of outbreaks, however, can be difficult due to the diffuse geographic and temporal distribution of the cases (Adak et al. 2005, Gilpin et al. 2006). The temporal association of cases can remain unclear, because the incubation period prior the onset of symptoms can vary. In addition, wide variation in the severity of the disease among individual patients complicates the detection of outbreaks. For example, patients with mild symptoms may recover without the need for medical care, and therefore remain unidentified as outbreak cases (Olson et al. 2008). A marked seasonality is characteristic to the incidence of human campylobacteriosis, which peaks in different summer months depending on the geographical area (Nylen et al. 2002, Kovats et al. 2005, Louis et al. 2005, Baker et al. 2007, van Hees et al. 2007, Ragimbeau et al. 2008, White et al. 2009). In the Nordic countries, for example, the number of human cases consistently peaks in the end of July and in the beginning of August (Nylen et al. 2002, Jore et al. 2010), whereas in England and Wales the peak occurs in mid-June and mid-July (Louis et al. 2005). The annual increase in the incidence of sporadic infections relates to climatic factors, such as rising ambient temperature (Patrick et al. 2004, Lake et al. 2009, Stark et al. 2009, White et al. 2009, Jore et al. 2010) and relative humidity (Patrick et al. 2004, White et al. 2009), whereas the effect of rainfall appears to be negligible (Patrick et al. 2004, Kovats et al. 2005, Louis et al. 2005). A

12

study by Nicholson et al. (2005), however, showed a significant association between preceding rainfall and water-borne outbreaks. Reports from different countries present the highest incidence rates among children under five years of age and in age groups between 15 and 29 years of age (Sopwith et al. 2003, Carrique-Mas et al. 2005, Baker et al. 2007, White et al. 2009, Nakari et al. 2010). Children living in rural areas seem to be at especially higher risk for contracting campylobacteriosis than those living in urban centres (Ethelberg et al. 2005, Baker et al. 2007, Garrett et al. 2007). Moreover, the incidence of campylobacteriosis in children under five years of age appears to be particularly temperature-related (Louis et al. 2005). Other factors, such as the use of acid-suppressing medication or underlying disease may also explain the higher risk of campylobacteriosis among the elderly reported in recent studies (Gillespie et al. 2009, Doorduyn et al. 2010). Besides differences among age groups, the incidence of campylobacteriosis also varies between genders. Males represent a slightly higher proportion of reported cases irrespective of age (Louis et al. 2005, Baker et al. 2007, White et al. 2009). Evidence from various studies has suggested that the development of immunity is a consequence of repeated or long-term exposure to Campylobacter spp., such as the regular consumption of risky food or occupational contact with animals (Forbes et al. 2009, Tam et al. 2009). Recent experiments with human volunteers have confirmed the acquisition of immunity, which offered complete short-term protection from illness, and resistance to colonisation upon re-challenge with the same C. jejuni strain (Tribble et al. 2010).

2.1.2 Sources of Campylobacter spp. in human infection

The predominantly sporadic appearance of campylobacteriosis complicates the tracing of its sources, which in sporadic cases often remain unidentified, because the incubation period prior to the onset of symptoms can be long. Nevertheless, in sporadic foodborne cases, a major source of campylobacters appears to be the handling and consumption of fresh chicken (Studahl and Andersson 2000, Adak et al. 2005, Wingstrand et al. 2006, Stafford et al. 2007, Unicomb et al. 2008, Wilson et al. 2008, Lindmark et al. 2009). More important than eating improperly heated chicken meat, however, is probably cross-contamination from raw chicken meat during meal preparation (Kapperud et al. 2003). The importance of chicken is obvious in countries such as Belgium, Iceland, Denmark and New Zealand, where the reduced consumption of chicken meat or the implementation of measures that reduce the contamination of chicken meat have substantially reduced the incidence of human cases (Vellinga and Van Loock 2002, Stern et al. 2003, Mullner et al. 2009, Rosenquist et al. 2009). On the other hand, the numbers of reported human cases have risen in Sweden and Finland despite the steady or reduced prevalence of campylobacters in chicken flocks (Studahl and Andersson 2000, EFSA 2010b). Moreover, genotyping studies of Campylobacter spp.

13

from different sources suggest overestimation of the importance of chicken in human campylobacteriosis (Duim et al. 2000, Dingle et al. 2001, Kärenlampi et al. 2003, Levesque et al. 2008). Besides the consumption and handling of chicken, case-control studies have identified other food-associated risk factors, including the consumption of undercooked meat, pork, pork with bones, ham and beef, offal, game and tripe, barbecued meat or undercooked seafood; eating at a restaurant; poor kitchen hygiene, and drinking unpasteurised or bird-pecked milk (Studahl and Andersson 2000, Kapperud et al. 2003, Neimann et al. 2003, Sopwith et al. 2003, Schönberg-Norio et al. 2004, Carrique-Mas et al. 2005, Gallay et al. 2006, Stafford et al. 2008, Unicomb et al. 2008, Doorduyn et al. 2010). In addition, the preparation of meat by barbecuing appears to be a risk factor for campylobacteriosis (Studahl and Andersson 2000, Kapperud et al. 2003, Neimann et al. 2003, Doorduyn et al. 2010). “Protective” food-related factors, in contrast, include for example the consumption of sausage, fish, raw vegetables, fruits or berries, chocolate and nuts and pasteurised milk (Kapperud et al. 2003, Schönberg-Norio et al. 2004, Carrique-Mas et al. 2005, Stafford et al. 2008, Doorduyn et al. 2010). Studies focusing on defined temporal and spatial areas have elucidated the relative importance of different sources of campylobacters in human infection. Increasing evidence from recent research indicates that exposures in urban areas differ from those in rural areas (Studahl and Andersson 2000, Baker et al. 2007, Garrett et al. 2007, Strachan et al. 2009). Poultry appears to be a less likely source of campylobacters among the rural population than among urban dwellers (Ethelberg et al. 2005, Mullner et al. 2010b). Moreover, a significant correlation between agricultural activities and the seasonality of infections in rural areas suggests an association with environmental rather than food sources (Kovats et al. 2005, Louis et al. 2005, Tam et al. 2006). The contaminated environment, direct contact with farm animals and the consumption of unpasteurised milk on the farm may be the most important exposures for rural population, and especially for children (Studahl and Andersson 2000, Kapperud et al. 2003, Sopwith et al. 2003, Minihan et al. 2004, Ethelberg et al. 2005, Schildt et al. 2006, Baker et al. 2007, Garrett et al. 2007, Strachan et al. 2009, Mullner et al. 2010b). In addition to food production animals, pet animals - especially young dogs and cats - can be carriers of thermophilic campylobacters (Hald and Madsen 1997, Hald et al. 2004, Wieland et al. 2005, Workman et al. 2005). Several studies have identified contact with dogs and cats as a risk factor for sporadic human campylobacteriosis (Kapperud et al. 2003, Neimann et al. 2003, Unicomb et al. 2008, Tam et al. 2009), particularly among infants (Carrique-Mas et al. 2005, Stafford et al. 2008, Doorduyn et al. 2010). Comparisons of genotypes of C. jejuni isolates from wildlife and the environment have yielded contradictory conclusions about the

14

importance of wild animals as an origin of human campylobacteriosis. Common C. jejuni genotypes in human disease occur in wildlife, such as birds (Colles et al. 2003, French et al. 2005), whereas other studies identify predominant subtypes from wild animals and the environment as a minor source of Campylobacter spp. in human infections (Broman et al. 2002, Colles et al. 2003, Broman et al. 2004, French et al. 2005, Garrett et al. 2007, Wilson et al. 2008). However, a major problem in studies of environmental campylobacters is the large diversity of inputs, so the environmental sampling may only provide an indication of the diversity of isolates present (Garrett et al. 2007). Several case-control studies have recognised the consumption of undisinfected water from a surface water source or a private well as a risk factor and, accordingly, the consumption of treated water as a “protective” factor against human sporadic campylobacteriosis (Kapperud et al. 2003, Neimann et al. 2003, Michaud et al. 2004, Nygård et al. 2004, Schönberg-Norio et al. 2004, Carrique-Mas et al. 2005, Sandberg et al. 2006). Furthermore, the largest outbreaks of campylobacteriosis have been water-borne and have often occurred as a consequence of contamination of drinking water supplies due to the washing out of faecal material of farm animals or wild birds from the environment after a heavy rain (Clark et al. 2003, Hänninen et al. 2003, Gallay et al. 2006, Pitkänen et al. 2008). Similarly, rainfall and the subsequent run-off can contaminate surface waters used for recreational purposes. Recently, recreational water exposure has appeared to be a risk factor in case-control studies (Schönberg-Norio et al. 2004, Denno et al. 2009, Doorduyn et al. 2010).

2.2 Subtyping of Campylobacter jejuni

The control of human campylobacteriosis requires a thorough understanding of the epidemiology of campylobacters. The special characteristics of these organisms, such as high diversity, weak clonality, frequent recombination within the genus, wide host distribution, and the sporadic nature of the disease, complicate the tracing the sources of these pathogens (Wassenaar and Newell 2000, Dingle et al. 2001, Strachan et al. 2009). Subtyping beyond the species level is therefore fundamental in gathering information on the relative importance of different sources in human campylobacteriosis from outbreak investigations, source attribution studies, and studies on the population genetics of pathogenic bacteria.

2.2.1 Serotyping

Serotyping is a traditional phenotypic subtyping method for epidemiological studies of C. jejuni and C. coli. Two serotyping schemes based on different antigens are available. The Penner serotyping scheme exploits the passive hemagglutination of heat-stable antigens of campylobacters (Penner and Hennessy 1980, Penner et al.

15

1983), later identified as capsular polysaccharides (Karlyshev et al. 2000), whereas the Lior scheme uses bacterial heat-labile antigens and slide agglutination (Lior et al. 1982). These previously widely used methods offer relatively low discriminatory power (Garrett et al. 2007, Gilpin et al. 2008b), and a high proportion of strains remains untypeable (Rautelin and Hänninen 1999, Desai et al. 2001, Devane et al. 2005). Therefore, either serotyping technique alone is ineffective as subtyping method. Additional disadvantages of serotyping include the limited commercial availability, high cost and poor quality of the antisera (Rautelin and Hänninen 1999, Desai et al. 2001).

2.2.2 Pulsed-field gel electrophoresis (PFGE)

PFGE is based on restriction site polymorphism throughout the entire genome using rare-cutting endonucleases. Immobilisation of the bacterial suspension in agarose plugs prior to cell lysis prevents the mechanical breakage of the genomic DNA (Wassenaar and Newell 2000). The genomic fragments (20 to 200 bp) are separated on agarose gel under particular conditions of electrophoresis in which the orientation of the electric field changes in a pulsed manner (Lukinmaa et al. 2004). The most commonly used restriction enzyme in PFGE for Campylobacter spp. is SmaI, which produces profiles that are sufficient to demonstrate the dissimilarity of isolates. However, demonstrating the similarity of isolates requires the use of two enzymes in digestion (Lindmark et al. 2004, Gilpin et al. 2006). Some studies have shown that digestion with KpnI alone is almost as discriminatory as the combination of SmaI and KpnI (Michaud et al. 2001, Gilpin et al. 2006). However, the reproducibility of results obtained with KpnI digestion appears to be poorer than those obtained with SmaI (Gilpin et al. 2006), which offers high reproducibility under standardised conditions (Ribot et al. 2001). The discriminatory power of PFGE is high (Hänninen et al. 2001, Sails et al. 2003). Variation among PFGE patterns arises from chromosomal insertions, deletions and recombination, which increases the discriminatory power of the method and its ability to detect rapidly occurring chromosomal changes (Levesque et al. 2008). Consequently, PFGE is a useful tool in focused short-term epidemiological studies, such as outbreak investigations, whereas it is less suitable for long-term longitudinal studies of epidemiology of campylobacters due to the wide genetic variability of these organisms (Engberg et al. 1998, Sails et al. 2003). The interpretation of PFGE patterns is, despite computer-aided analysis methods, based largely on the subjective visual comparison of profiles. The lack of standardisation limits comparisons of typing results among different laboratories. The protocols of Pulsenet (Ribot et al. 2001) and Campynet (http://campynet.vetinst.dk ) are attempts towards harmonisation of this genotyping method.

16

2.2.3 Amplified fragment length polymorphism (AFLP)

AFLP method is based on the selective amplification of chromosomal DNA fragments obtained by the use of two restriction endonucleases. After digestion of DNA and the subsequent ligation of restriction site-specific adapters and preselective PCR, the final selective amplification of DNA fragments with radioactively or fluorescently labelled primers results in products from 50 to 500 bp. The final PCR products are separated on denaturing polyacrylamide gels and analysed using an automated sequencer (Duim et al. 1999). AFLP is a highly discriminatory subtyping method (de Boer et al. 2000, Hänninen et al. 2001), which appears to be less sensitive than PFGE to the genetic instability (Wassenaar and Newell 2000) However, the cost of the equipment and the difficulty of making interlaboratory comparisons are major disadvantages of this method (Wassenaar and Newell 2000, Schouls et al. 2003).

2.2.4 Fla-SVR typing

Fla-SVR typing is a technique which uses PCR amplification of the short variable region (SVR) of the flaA flagellin gene for sequencing. This region, although short (321 bp), is hypervariable and can discriminate even closely related campylobacter strains (Meinersmann et al. 1997, Dingle et al. 2001, Meinersmann et al. 2005); the technique is therefore valuable in outbreak investigations (Sails et al. 2003). However, the flaA locus may be unsuitable for longitudinal epidemiological studies due to intra- and intergenomic recombination (Harrington et al. 1997, Sails et al. 2003). The major advantage of this method, like other sequence-based methods, is the objective interpretation and standardised nomenclature of the subtypes which permit interlaboratory comparisons and electronic distribution (Sails et al. 2003)

2.2.5 Multilocus sequence typing (MLST)

Multilocus sequence typing utilises the genetic variation of the nucleotide sequences of ca. 500-bp fragments from seven housekeeping genes, which are slowly evolving as they are essential to metabolic function (Dingle et al. 2001, Wareing et al. 2003). Using the nucleotide sequence data, isolates can be assigned a sequence type (ST), which represents a combination of seven numbers obtained by assigning a number to each unique allele at a specific locus. This typing method allows the examination of the population structure of campylobacters in terms of clonal complexes. Each clonal complex, representing a lineage believed to originate from a common ancestor, consists of a central genotype, a founder ST, after which the complex is named, together with closely related genotypes. Generally, the

17

founder represents a frequently occurring genotype, whereas the other members of the clonal complex are less common (Dingle et al. 2001, Wareing et al. 2003). MLST was developed to be a tool in studies of population genetics and evolutionary studies (Dingle et al. 2001, Wareing et al. 2003), and is especially suitable for identification of clonal complexes among genetically diverse bacterial species such as C. jejuni (Wareing et al. 2003). Due to the wide geographical distribution of sequence types or clonal complexes, MLST is an especially an invaluable tool for long-range epidemiological studies (Dingle et al. 2008). The discriminatory power of MLST is comparable to that of flaA SVR typing (Levesque et al. 2008). However, MLST is less discriminatory than PFGE, and is therefore less suitable for outbreak investigations (Sails et al. 2003, Levesque et al. 2008). The applicability of MLST to short-term epidemiological studies increases when additional loci, such as the flaA SVR or nucleotide sequences of genes encoding antigens, are included in the analysis (Sails et al. 2003, Dingle et al. 2008). The advantages of the method are its objectivity, reproducibility and simplicity of interpretation of the results (Dingle et al. 2001). As a sequence-based typing method, MLST is portable, and the sequence data are comparable between laboratories due to its unified nomenclature (McCarthy et al. 2007, Levesque et al. 2008). A freely accessible international database of Campylobacter MLST data is available (http://mlst.zoo.ox.ac.uk).

2.2.6 DNA microarray

Microarray technology enables comparisons of DNA from whole bacterial genome sequences, and, in combination with sophisticated mathematical algorithms, permits the determination of phylogenetic relationships between bacterial populations. Comparative phylogenetics provides an approach to investigate differences in the genomes of isolates from different sources and to identify specific genes associated with particular animal hosts or with the virulence of pathogenic bacteria (Dorrell et al. 2002, Taboada et al. 2004, Champion et al. 2005). In studies on the comparative phylogenetics of C. jejuni, the genomic sequence of pathogenic isolate NCTC 11168 (Parkhill et al. 2000) is the reference strain most commonly used as the basis of whole-genome DNA microarrays (Champion et al. 2005). The exploitation of the complete genome data is a definite advantage of this approach in comparison to other subtyping methods (Taboada et al. 2004). However, a disadvantage is its use of only a single reference strain, which may exclude a fraction of the gene pool of C. jejuni (Champion et al. 2005).

18

2.3 Campylobacter spp. in cattle

2.3.1 Prevalence of Campylobacter spp. in cattle

Thermophilic campylobacters are typically the most frequently isolated human bacterial pathogens from healthy cattle at slaughter (Beach et al. 2002, Gharst et al. 2006, Madden et al. 2007, Milnes et al. 2008). In slaughterhouse surveys, the prevalence of bovine intestinal campylobacter colonisation has varied between 12.5% and 89.4% (Table 1). Furthermore, studies on campylobacters on cattle farms or in cattle herds have reported percentages from 12% to 100% (Busato et al. 1999, Wesley et al. 2000, Nielsen 2002, Englen et al. 2007, Oporto et al. 2007, Parisi et al. 2007, Gilpin et al. 2008b, Kwan et al. 2008b, Ragimbeau et al. 2008, Ellis-Iversen et al. 2009a), and within-herd prevalences from 0% to 100% in dairy cattle (Humphrey and Beckett 1987, Oporto et al. 2007, Gilpin et al. 2008a, Gilpin et al. 2008b, Pradhan et al. 2009), and from 5.4% to 83% in beef cattle (Inglis et al. 2003, Berry et al. 2006, Oporto et al. 2007). However, the results of different studies are not fully comparable due to variations in study designs and laboratory methods. The intestinal sampling site in slaughterhouse surveys (Garcia et al. 1985, Grau 1988, Stanley et al. 1998, Inglis et al. 2005), the sampling methods on farms (Hoar et al. 1999), the age of animals (Nielsen 2002) and the detection methods in the laboratory (Stanley et al. 1998, Inglis et al. 2003, Gharst et al. 2006) all influence the results.

19

Table 1. Prevalences of thermophilic Campylobacter spp. in cattle in slaughterhouse surveys.

Sample type No. of animals

examined Proportion of

positive samples, %

Reference

Rumen, calves Faeces, calves Rumen, adult cattle Faeces, adult cattle

23 24 89 96

74 54 3.4

12.5

Grau 1988

Gallbladder Large intestine Small intestine Liver Lymph node

100 100 100 100 70

33 35 31 12 1.4

Garcia et al. 1985

Intestinal contents

360

89.4

Stanley et al. 1998

Rectal swab, feedlot cattle Rectal swab, adult cattle

100 96

68 7

Beach et al. 2002

Gallbladder, intestinal contents, liver or faeces

1154

26.1

Acik and Cetinkaya 2005

Faeces, beef cattle Faeces, dairy cattle

252 358

19 95

Gharst et al. 2006

Intestinal contents, calves Intestinal contents, adult cattle

74 715

46

28.5

Johnsen et al. 2006

Faeces, beef cattle

220

24.8

Madden et al. 2007

Rectal contents (1999/2000) Rectal contents (2003)

667 891

54.6 24.5

Milnes et al. 2008

Liver Bile

108 108

45 5

Enokimoto et al. 2007

Liver

60

31.7

Ghafir et al. 2007

Bile Liver

290 148

23 1.4

Matsumoto et al. 2008

Faeces, calves Faeces, beef cattle Faeces, culled cows

747 754 754

39.1 6.0 4.6

Chatre et al. 2010

20

2.3.2 Campylobacter species in cattle

C. jejuni has been predominant, whereas C. coli has become a minor species in cattle in most of the slaughterhouse and farm studies (Garcia et al. 1985, Giacoboni et al. 1993, Stanley et al. 1998, Wesley et al. 2000, Minihan et al. 2004, Acik and Cetinkaya 2005, Bae et al. 2005, Berry et al. 2006, Madden et al. 2007, Oporto et al. 2007, Parisi et al. 2007, Gilpin et al. 2008a, Gilpin et al. 2008b, Milnes et al. 2008, Ragimbeau et al. 2008, Ellis-Iversen et al. 2009a, Chatre et al. 2010). Beside these well-known human pathogens, other Campylobacter spp. of unclear importance to human health appear to be common in bovine intestines. Some surveys have identified C. hyointestinalis (Grau 1988, Atabay and Corry 1998, Pezzotti et al. 2003), and a new species, C. lanienae, as the most prevalent Campylobacter species in the intestines of cattle (Inglis et al. 2003, Inglis and Kalischuk 2004, Inglis et al. 2004). A minor bovine intestinal species is C. fetus (Giacoboni et al. 1993, Atabay and Corry 1998, Busato et al. 1999, Inglis et al. 2003, Inglis et al. 2004), the two subspecies of which cause genital infections and abortions in cattle and can infect immunodeficient humans (Debruyne et al. 2008). Co-colonisation of at least two Campylobacterspp. can occur in cattle faeces (Inglis et al. 2003, Inglis et al. 2004) or in the gallbladder (Enokimoto et al. 2007). An animal’s age appears to influence to the proportions of different Campylobacter spp. present in the faeces of cattle (Giacoboni et al. 1993, Busato et al. 1999, Bae et al. 2005).

2.3.3 Campylobacter jejuni in cattle at farm

Cattle are usually symptomless carriers of campylobacters (Stanley et al. 1998). However, C. jejuni can cause diarrhoea - sometimes with severe symptoms - in young cattle (Dilworth et al. 1988, Gilpin et al. 2008b). Although free of campylobacters at birth, calves acquire these organisms in an early phase of life due to exposure to a contaminated environment (Stanley et al. 1998, Gilpin et al. 2008b), and are more frequent carriers of campylobacters than adult cattle (Giacoboni et al. 1993, Nielsen 2002, Johnsen et al. 2006, Gilpin et al. 2008b, Chatre et al. 2010). In addition, calves excrete higher numbers of campylobacters in their faeces than do older animals (Stanley et al. 1998, Nielsen 2002), although the diversity of C. jejuni subtypes in adult cattle may be greater (Nielsen 2002, Kwan et al. 2008b). Studies of the shedding patterns of C. jejuni in cattle herds have reported that individual animals can be persistent carriers and shedders of high numbers of C. jejuni or even of a single subtype of C. jejuni, whereas others excrete Campylobacter spp. intermittently (Humphrey and Beckett 1987, Hänninen et al. 1998, Stanley et al. 1998, Inglis et al. 2004, Minihan et al. 2004, Gilpin et al. 2008b, Kwan et al. 2008b).

21

Nevertheless, some individuals appear to be resistant to colonisation in an environment where the exposure rate is high (Minihan et al. 2004). The variety of environmental sources of C. jejuni is great, when the cattle are grazing outdoors (Oporto et al. 2007, Grove-White et al. 2010). For example, one farmland study detected an association between the presence of C. jejuni in bird faeces and a higher probability of isolating the organism from cattle (Brown et al. 2004). On the other hand, indoor housing can allow re-infection from a faecally contaminated environment or due to closer contacts with carriers of Campylobacter spp. (Stanley et al. 1998, Busato et al. 1999, Minihan et al. 2004, Ellis-Iversen et al. 2009a, Ellis-Iversen et al. 2009b). Large herd size, which can relate to higher stocking density of cattle, is likely to increase contact between animals and appears to be a risk factor for faecal shedding of Campylobacter spp. (Ellis-Iversen et al. 2009b, Grove-White et al. 2010). An important factor in the transmission of campylobacters among cattle is drinking water hygiene. Water from private supplies appears to be a risk factor for colonisation of Campylobacter spp. in young cattle (Ellis-Iversen et al. 2009b). In addition, campylobacter contamination of water trough surfaces appears to increase (Minihan et al. 2004), and, unsurprisingly, the frequent emptying and cleaning of water troughs reduces the risk for campylobacter infection (Ellis-Iversen et al. 2009a). Without cleaning, the chlorination of drinking water alone seems insufficient to prevent transmission of the organism among cattle reared indoors (Wesley et al. 2000, Besser et al. 2005). During the grazing period, campylobacter colonisation may persist due to the cattle’s access to natural waters (Humphrey and Beckett 1987, Hänninen et al. 1998). A strong seasonal fluctuation in the occurrence of Campylobacter spp. is evident in dairy cattle with highest prevalences occurring in late spring or summer when the cattle are grazing (Hänninen et al. 1998, Stanley et al. 1998, Kwan et al. 2008b, Grove-White et al. 2010). Besides the water source, changes in diet can affect the colonisation and shedding of campylobacters in cattle at pasture (Stanley et al. 1998, Ellis-Iversen et al. 2009b, Grove-White et al. 2010). In addition, the presence of wildlife may increase the exposure of cattle to campylobacters, whereas direct transmission between individuals in a herd may occur less frequently than when animals are housed indoors (Grove-White et al. 2010). The transmission of Campylobacter spp. from other production animals, such as pigs, on the same farm can occur at low levels (Boes et al. 2005): one study has identified the presence of horses as a risk factor for the campylobacter colonisation of young cattle (Ellis-Iversen et al. 2009b). Other factors that may increase the risk for campylobacter colonisation of cattle include the type of feed, manure disposal on the farm, the accessibility of feed to wild birds (Wesley et al. 2000), the effects of reproductive hormones (Stanley et al. 1998), or metabolic stress due to the demands on production animals (Grove-White et al. 2010). Among intensively raised feedlot cattle, for

22

example, the faecal shedding of Campylobacter spp. can substantially increase during the relatively short feeding period (Minihan et al. 2004, Besser et al. 2005).

2.3.4 Genetic diversity and host adaptation of bovine Campylobacter jejuni strains

C. jejuni strains isolated from cattle represent a wide variety of genotypes. Farm studies have identified as many as nine different genotypes simultaneously present in a herd (Nielsen 2002, Oporto et al. 2007, Parisi et al. 2007, Gilpin et al. 2008a, Ragimbeau et al. 2008), and co-colonisation of two or more non-related C. jejuni genotypes in one animal has also occurred (Gilpin et al. 2008a, Gilpin et al. 2008b). The diversity of campylobacter genotypes in cattle may reflect the number of various sources of these organisms due to different farming practices (Nielsen 2002, Parisi et al. 2007), although it may also indicate that the bovine intestinal tract is a favourable environment for the exchange of genetic material among campylobacter strains (French et al. 2005, Meinersmann et al. 2005, McCarthy et al. 2007). Through intragenetic or intergenetic recombination, C. jejuni can adapt to persistent colonisation in the intestines of a specific host and acquire a host signature in the genome, which can predict the source of the organism in human infections (Dingle et al. 2001, Champion et al. 2005, McCarthy et al. 2007). An example of cattle- or ruminant-associated genotypes is the C. jejuni ST-61 clonal complex, which, according to reports from a few countries in Europe and from New Zealand, occurs predominantly in cattle (Colles et al. 2003, Manning et al. 2003, French et al. 2005, Kärenlampi et al. 2007, Kwan et al. 2008b, Ragimbeau et al. 2008, Mullner et al. 2010a). Evidence from MLST studies suggests that this clonal complex of C. jejuni has evolved in the intestines of cattle and other ruminants, and that the particular allele (uncA17) which defines the ST-61 likely originates from C. coli (Dingle et al. 2002, French et al. 2005, Meinersmann et al. 2005).

2.4 Campylobacter spp. in foods of bovine origin

2.4.1 Beef and edible offal

Although cattle frequently carry campylobacters when arriving at the slaughterhouse, (Besser et al. 2005, Garrett et al. 2007), red meat appears to be a minor source of these organisms (Table 2). The faecal campylobacter contamination of carcasses is possible during processing, but a high-level slaughter hygiene reduces overall contamination (Minihan et al. 2004, Garrett et al. 2007), and drying, along with exposure to oxygen during chilling further decreases the survival of Campylobacter spp. on carcasses and in red meat (Grau 1988). Minced meat, rather, can provide favourable conditions for the

23

survival of campylobacters at refrigerator temperature (Svedhem et al. 1981). However, studies on ground beef at retail have typically failed to detect Campylobacter spp. (Ghafir et al. 2007, Medeiros et al. 2008, Phillips et al. 2008).

Table 2. Occurrence of thermophilic Campylobacter spp. in retail beef

Total No. of samples

No. of positive samples

Proportion of positive samples, %

Reference

182 1 0.5 (Zhao et al. 2001)

151 2 1.3 (Pezzotti et al. 2003)

221 7 3.2 (Whyte et al. 2004)

230 8 3.5 (Wong et al. 2007)

250 3 1.2 (Hong et al. 2007)

451 49 10.9 (Hussain et al. 2007)

50 1 2.0 (Vindigni et al. 2007)

1514 71 4.7 (Little et al. 2008)

198 22 11.1 (Bostan et al. 2009)

210 5 2.4 (Rahimi et al. 2010)

142 20 14.1 (Sammarco et al. 2010)

Apparently healthy cattle may carry Campylobacter spp. in the gallbladder (Garcia et al. 1985, Enokimoto et al. 2007). Bile can therefore transmit campylobacter contamination to the liver during the slaughter process (Acik and Cetinkaya 2005, Enokimoto et al. 2007, Little et al. 2008, Matsumoto et al. 2008). Surveys at slaughter have reported campylobacter prevalences between 1.4% and 45% (Table 1), and retail studies have presented prevalences of 12% to 54% in the liver (Kramer et al. 2000, Little et al. 2008, Medeiros et al. 2008).

24

2.4.2 Milk and milk products

The common presence of Campylobacter spp. in the intestines of dairy cattle warrants the possibility of faecal contamination of raw milk. The contamination of milk can occur due to lapses in hygiene or failures in the milking process, but can be avoided or at least reduced by applying proper hygiene at milking, and pasteurising milk, which destroys campylobacters (Humphrey et al. 2007). The prevalences ofCampylobacter spp. in raw milk have varied from 0% to 27% (Table 3), and concentrations from lower than 10 cfu/ml up to 100MPN/100 ml (Humphrey and Beckett 1987, Heuvelink et al. 2009). Few studies have explored the presence and survival of Campylobacter spp. in milk products. The preparation processes of Brie and Camembert cheeses or hard and semi-hard cheeses seem unfavourable to campylobacters (Bachmann and Spahr 1995, Medeiros et al. 2008), and the survival of C. jejuni in yoghurt is poor (Birk and Knochel 2009), probably due to low pH, and the presence of organic acids and other metabolites produced by lactic acid bacteria. C. jejuni, however, was able to survive up to 18 days in garlic butter at refrigerator temperature when the initial inoculum was large (Zhao et al. 2000).

Table 3. Prevalence of Campylobacter spp. in raw milk in different studies

Total No. of samples

No. of positive samples

Proportion of positive samples, %

Reference

108 1 0.9 (Doyle and Roman 1982)

210 3 1.4 (Lovett et al. 1983)

111 9 8.1 (Humphrey and Beckett 1987)

111 1 0.9 (Hudson et al. 1999)

131 12 9.2 (Jayarao and Henning 2001)

300 82 27.3 (Yang et al. 2003)

62 1 1.6 (Whyte et al. 2004)

248 5 2.2 (Jayarao et al. 2006)

127 13 10.2 (Hussain et al. 2007)

59 0 0 (Medeiros et al. 2008)

25

2.5 Cattle as a source of Campylobacter spp. in human infections

2.5.1 Outbreak investigations and case-control studies

Investigations have attributed numerous outbreaks of campylobacteriosis to the consumption of unpasteurised or improperly pasteurised milk, or of products prepared from unpasteurised milk (Robinson et al. 1979, Morgan et al. 1994, Fahey et al. 1995, Lehner et al. 2000, Peterson 2003, Centers for Disease Control and Prevention (CDC) 2009, Heuvelink et al. 2009, Unicomb et al. 2009). The consumption of raw milk during farm visits (Evans et al. 1996, Kalman et al. 2000), in camps (McNaughton et al. 1982, Lehner et al. 2000), festivals (Morgan et al. 1994) schools or day-care centres (Jones et al. 1981, Robinson and Jones 1981) has resulted in wide outbreaks in many countries. In Finland, the faecal contamination of milk due to a failure in the milking process caused a long-lasting outbreak of campylobacteriosis among members of a farming family who consumed raw milk (Schildt et al. 2006). Reports from case-control studies have also identified the consumption of unpasteurised milk as an important risk factor for campylobacteriosis among humans (Studahl and Andersson 2000, Kapperud et al. 2003, Neimann et al. 2003, Michaud et al. 2004), especially among children (Carrique-Mas et al. 2005). Other risk factors related to food of bovine origin include the consumption of steak tartare (a raw beef product) (Doorduyn et al. 2010) or barbecued red meat (Neimann et al. 2003). Several recent case-control studies have examined the different risks of campylobacteriosis available in rural and urban areas. In a Danish study, the risk for infection appeared higher among people -particucarly children - living in areas of low population density or in farm houses than in urban–type housing (Ethelberg et al. 2005), and another study reported rising campylobacteriosis incidence associated with increasing ruminant density in Sweden (Nygård et al. 2004). In Walkerton, Canada, campylobacters originating from neighbouring cattle farms contaminated the municipal water supply after a heavy rain and caused a large-scale water outbreak (Clark et al. 2003). Furthermore, an increased risk for campylobacteriosis has been associated with contact with cattle (Kapperud et al. 2003, Neimann et al. 2003), or with farm animals more generally, including cattle (Michaud et al. 2004, Doorduyn et al. 2010). Indeed, direct contact with diarrhoeic calves or with bovine faecal material appeared to be the cause of a campylobacter infection of farm workers or children living on farm (Dilworth et al. 1988, Gilpin et al. 2008a).

26

2.5.2 Genotyping and source attribution studies

Subtyping campylobacters enables the attribution of human infections to specific sources. Source attribution studies that have compared campylobacter isolates from human infections and from potential sources of infection, and examined risk factors related to specific C.jejuni genotypes have provided additional information about the role of cattle in human campylobacteriosis. PFGE of human and cattle isolates in temporally and spatially defined studies has shown genotypic similarities indicating cattle as potential source of Campylobacter spp. in humans (Fitzgerald et al. 2001, Devane et al. 2005, Johnsen et al. 2006, Garrett et al. 2007, Gilpin et al. 2008b). Comparisons of human and animal isolates from various collections representing several time periods have indicated that some ST complexes, especially ST-61 complex, commonly isolated in human infections (Dingle et al. 2001) are unexpectedly common in cattle (Dingle et al. 2002, Manning et al. 2003, Schouls et al. 2003, Kärenlampi et al. 2007). Recent spatio-temporally focused MLST studies in the farm environment have confirmed the association of ST-61 complex with cattle (French et al. 2005, Kwan et al. 2008a), and have identified additional bovine-adapted STs occurring in humans as well (Rotariu et al. 2009, Sheppard et al. 2009). The importance of bovine sources was evident in a study reporting that 42% of campylobacter isolates from infections in young children in a rural area represented STs similar to those from cattle (Strachan et al. 2009). Cattle have become a potential origin of human campylobacter infections in genotype-specific risk factor studies. Human infections by C. jejuni STs associated with ruminants, especially among children in rural areas have been more frequent in rural areas than in urban areas (Mullner et al. 2010b). Furthermore, ST-48 (a sequence type occurring especially in cattle) in patients was associated with eating and tasting raw minced meat (Kärenlampi et al. 2007), and a certain flaA subtype (the third most common type in human infections in Australia) was associated with the consumption of undercooked beef (Unicomb et al. 2008).

2.6 Antimicrobial resistance of Campylobacter spp.

As a self-limiting disease, campylobacter enteritis rarely requires antimicrobial therapy, which may, however, be necessary for patients with severe symptoms, prolonged duration of the infection, or an underlying disease (Blaser and Engberg 2008). The first choice of treatment of the disease is the macrolides: erythromycin, clatrithromycin or azithromycin (Bywater et al. 2004, Gupta et al. 2004, Blaser and Engberg 2008). The previous practice of using fluoroquinolones, especially in travel-related enteric infections, may be ineffective in the treatment of campylobacteriosis due to the rapidly rising antimicrobial resistance of Campylobacter spp. (Gupta et al. 2004), which is common among the C. jejuni and C. coli isolates from

27

animals and food in several European countries (EFSA 2010a). The development of resistance to fluoroquinolones among campylobacters has occurred concurrently with the extensive use of these antimicrobials in food production animals (Endtz et al. 1991, Levesque et al. 2008), and the veterinary use of fluoroquinolones appears to be a plausible explanation for the increased resistance among Campylobacter spp. rather than their use in human medicine (Engberg et al. 2004). Fluoroquinolone treatment can, in rare occasions, induce the emergence of resistant strains in human patients (Wistrom and Norrby 1995). However, patients are insignificant as sources of resistant campylobacter strains due to the minor role of person-to-person transmission in the epidemiology of campylobacteriosis (Engberg et al. 2004). Nevertheless, with regard to human campylobacter infections, decreased susceptibility to antimicrobial agents among Campylobacter spp. is a major concern, because the range of antimicrobial agents available for the treatment of severe infections may be considerably compromised, and failures in treatment are possible (Anderson et al. 2001). Furthermore, evidence from some studies indicates that human infections caused by resistant campylobacter strains may be prolonged or become more serious then those caused by susceptible strains (Engberg et al. 2004, Gupta et al. 2004, Helms et al. 2005, Feodoroff et al. 2009). Recently, the WHO has defined fluoroquinolones and macrolides as critically important antimicrobials in human medicine (WHO 2007) and recommended urgent development of risk management strategies for maintaining the effectiveness of these agents. The determination of minimum inhibitory concentrations (MICs) is the recommended method for examination of the antimicrobial susceptibility of pathogenic bacteria (EUCAST 2003, CLSI 2008). To monitor the development of antimicrobial resistance, the European Food Safety Authority (EFSA 2007) recommends interpreting of the data according to epidemiological cut-off values (Table 4), which separate the wild-type bacterial population and isolates with reduced susceptibility to antimicrobial agents (Kahlmeter et al. 2003), instead of the clinical breakpoint values, which are the criteria in the therapeutic approach (Schwarz et al. 2010).

28

Table 4. Epidemiological cut-off values and clinical breakpoints of antimicrobial susceptibility for Campylobacter jejuni and C. coli

Epidemiological cut-off value, mg/la

Clinical breakpoint, mg/lb

Antimicrobial agent

C. jejuni C. coli C. jejuni/coli

Cloramphenicol >16 >16 NDc

Ciprofloxacin >1 >1 �4

Erythromycin >4 >16 �32

Gentamicin >1 >2 ND

Nalidixic acid >16 >32 ND

Streptomycin >2 >4 ND

Tetracycline >2 >2 �16

a EUCAST http://www.eucast.org/mic_distributions/b CLSI 2006 c Not determined for Campylobacter spp.

29

3 Aims of the study

The aim of this study was to investigate the role of Finnish cattle as a potential reservoir of thermophilic Campylobacter spp., and antibiotic-resistant Campylobacter jejuni, and as a source (besides chicken) of domestically acquired sporadic human campylobacteriosis in Finland. The specific objectives were: I. to determine the prevalence of thermophilic

Campylobacter spp. in Finnish cattle at slaughter as well as the diversity and antimicrobial susceptibility of bovine C. jejuni isolates.

II. to investigate the colonisation dynamics of C. jejuni in

three Finnish dairy cattle herds. III. to develop genetic markers for investigation of the host

association of C. jejuni strains isolated from cattle, chickens and humans.

IV. to evaluate the contributions of chickens and cattle as

sources of domestically acquired sporadic human C. jejuni infections in Finland in the summer of 2003.

30

4 Materials and methods

4.1 Sampling

In study I, 952 rectal faecal samples and 948 carcass surface samples were collected from 12 Finnish slaughterhouses from January to December 2003. The number of samples and the frequency of sampling were determined on the basis of the slaughter volumes of each slaughterhouse during the previous year. The faecal material from randomly chosen animals was collected into plastic sampling jars, leaving only a small air space in order to prevent the adverse effects of oxygen on the survival of the campylobacters. Carcass surface samples including the brisket, inner and outer thigh, and the pelvic cavity of the same animals, were taken using premoistened sterile gauze pads placed in sterile plastic bags for transportation. In study II, three campylobacter-positive dairy cattle herds (15, 20 and 90 animals) located 60 km apart from each other in Southern Finland, were sampled over a one-year period on five occasions: 1) after the grazing period in November 2006, 2) in the middle of winter housing period in January-February 2007, 3) before the new grazing period in April 2007, 4) during the grazing in August 2007, and 5) after the grazing period in November 2007. On each sampling occasion, between 17 and 33 samples of newly-avoided faeces from individual animals were collected from the floor. Animals recently treated with antimicrobials were excluded from the sampling. In addition, tank milk samples were taken on each occasion. During the last sampling, drinking troughs of the animals were sampled using sponge swabs (Medical Wire & Equipment, Corsham, Wiltshire, UK).

4.2 Isolation of Campylobacter spp. (I, II)

All faecal samples of the slaughterhouse survey (I) and the farm study (II) were examined using enrichment. Ten grams of faecal material were weighed into 90 ml of Bolton broth (Campylobacter Enrichment Broth, Lab 135 plus selective supplement X131 [LAB M, Bury, England] plus lysed horse blood). In study II, a 10-fold dilution series up to 10-6 in Bolton broth was cultured for the semiquantitative detection of Campylobacter spp. (NCFA [Nordic Committee on Food Analysis] 2007). Broth cultures were incubated at 41.5°C for 24 h in a microaerobic incubator (ThermoForma [Thermo Electron Corporation, Marietta, OH]) (O2, 5%; CO2, 10%; N2, 85%). One loopful (10 �l) of enrichment culture was spread onto modified Charcoal Cefoperazone Deoxycholate Agar (mCCDA) plates (Campylobacter Blood Free Selective Medium Lab 112 plus supplement X112 [LAB M, Bury,

31

England]), which were incubated under the same conditions. The gauze samples from carcasses (I) and the sponge swab samples from drinking troughs (II) were similarly enriched in 225 ml of Bolton broth. In study I, an additional 10-�l loopful of 730 faecal samples was directly cultured on mCCDA for comparison of the two detection methods. The most probable number (MPN) technique was applied to quantify Campylobacter spp. in the tank milk samples (II). Either 10×100 ml or 10×20 ml of raw milk was enriched in Bolton broth (100 ml of milk + 500 ml of Bolton broth or 20 ml of milk + 80 ml of Bolton broth). The enrichment cultures were incubated microaerobically at 37�C for 48 h and plated on mCCDA plates which were incubated microaerobically at 37�C for 48 h. A minimum of two typical colonies from each mCCDA plate were subcultured onto Brucella agar (BBL, Becton Dickinson, MD) supplemented with 5% whole bovine blood treated with sodium citrate. A minimum of two isolates per campylobacter-positive sample were biochemically identified to the species level according to the standard method ISO 10272-1:2006 (ISO 2006). H2S production in triple-sugar iron agar (TSI, pH 8) (LAB M, Bury, England) and the urease production of hippurate-negative, indoxyl acetate-hydrolysing isolates were examined to identify C. hyointestinalis strains. The isolates were stored in Brucella broth (BBL, Becton Dickinson, MD) supplemented with 15% glycerol at -70°C.

4.3 Campylobacter jejuni isolates (III, IV)

4.3.1 Human isolates (III, IV)

In study III, domestically acquired human C. jejuni isolates (n=309) were isolated in six local laboratories from July to September 1999 (Kärenlampi et al. 2003) and at the Helsinki University Central Hospital Laboratory throughout the year in 1996, 2002 and 2003 (Kärenlampi et al. 2007). Altogether 175 domestic human C. jejuni isolates, collected in nine clinical microbiology laboratories (Figure 1) across the country from June to August 2003 were included in study IV. The strains were isolated from faecal samples of diarrhoeic patients by direct culture on mCCDA. These laboratories submitted all domestic isolates to the National Public Health Institute (KTL; currently the National Institute for Health and Welfare [THL]) for further examination. An isolate was considered domestic if the patient had not travelled abroad within ten days prior to the onset of symptoms or within 17 days before the specimen was taken. Isolates from identified outbreaks were excluded.

32

4.3.2 Chicken isolates (III, IV)

The chicken C. jejuni isolates in study III represented all chicken slaughter batches from the three Finnish slaughterhouses in the summer of 1999 (Perko-Mäkelä et al. 2002) and retail chicken meat samples from the Helsinki area from July to September 2003 (Kärenlampi et al. 2007). Chicken C. jejuni isolates (n=43) represented all chicken batches (n=955) slaughtered between May and August 2003 in two of the three Finnish broiler slaughterhouses (IV) (Figure 1). The strains were isolated in slaughterhouse laboratories by direct culture on mCCDA of the caecal contents from three to five chickens per slaughter batch. One isolate from each campylobacter-positive slaughter batch was submitted to the Finnish Food Safety Authority (Evira) for further investigation.

4.3.3 Bovine isolates (III, IV)

The bovine C. jejuni isolates (n=131) in study III were selected from the isolates from bovine faeces in study I. In study IV, we compared all faecal C. jejuni isolates (n=186) collected in the cattle slaughterhouse survey (I) throughout the entire year to human domestic isolates collected during the seasonal peak, because we assumed that the herds from which the campylobacter -positive animals came continuously carried the same PFGE types (as occurred in the three herds in study II). Consequently, these types could infect humans during the summer. In addition, all carcass isolates (n=15) from sampling between May and August 2003 were included to represent possible transmission via beef.

4.4 Serotyping of Campylobacter jejuni isolates (I)

C. jejuni isolates from bovine faecal and carcass samples were serotyped using a set of 25 commercial antisera for the serotyping of heat-stable antigens (Penner) of C. jejuni using the passive hemagglutination method (Denka Seiken Co., Ltd., Tokyo, Japan). Tests were performed, and the results were interpreted according to the manufacturer’s instructions.

33

Figure 1. Location of clinical microbiology laboratories and chicken slaughterhouses included in study IV, and cattle farms in the slaughterhouse survey (I).

34

4.5 Pulsed-field gel electrophoresis (I, II, IV)

The agarose plugs for PFGE analysis were prepared according to the PulseNet protocol (www.cdc.gov/pulsenet/protocols, (Ribot et al. 2001) and stored in Tris-EDTA buffer at 4°C. The DNA was digested overnight at 25°C with 20 U of SmaI, or for a minimum of 4 h at 37°C with 20 U of KpnI restriction endonuclease (New England Biolabs Inc., Ipswich, MA) in a final volume of 200 �l with 2 �l of bovine serum albumin (New England Biolabs Inc., Ipswich, MA). An agarose gel (1%) was prepared in 0.5 × Tris-buffered EDTA (Sigma-Aldrich Co, Baltimore, MD). Fragments were separated by electrophoresis for 18 h at 6 V and 14°C with ramped pulse times from 6.8 to 35.4 s with a CHEF-DRIII pulsed-field electrophoresis system (Bio-Rad, CA). The gels were stained for 45 min with ethidium bromide (0.5 �g/ml) and photographed under UV light. The PFGE data were analysed with Bionumerics V5.10 (Applied Maths, Kortrijk, Belgium) at 0.5% optimisation and 1.0% tolerance. The PFGE pattern of Salmonella Braenderup H9812 (ATCC BAA-664) served as the fragment size marker. Profiles differing by one or more bands were considered different subtypes. The criteria presented by (Tenover et al. 1995) were applied to assess the relationship of the subtypes (I).

4.6 PCR of genetic markers of Campylobacter jejuni (III)

Four genetic markers were selected from the completely sequenced genomes of C. jejuni strains 81-176 (Hofreuter et al. 2006), RM1221, and NCTC 11168 using comparative genomics (Chaudhuri et al. 2008), and primers were designed for the detection of these markers, which were ggt, the �-glutamyl transpeptidase gene; dmsA (Cju34), a subunit of the putative tripartite anaerobic dimethyl sulfoxide (DMSO) oxidoreductase (DMSO/trimethylamine N-oxide reductase) gene; Cj1585c, coding for a putative oxidoreductase; and CJJ81176-1371, a putative serine protease gene. The presence of these four genes in C. jejuni isolates from bovine faecal samples (n=131), chicken caecal or meat samples (n=205), and human patients (n=309) was examined using PCR to assess their applicability for host association studies. PCR primers designed for the amplification of the fragments appear in Table 5. Twelve PCR products for each gene fragment were sequenced to find the similarity of the sequences within a gene.

35

Table 5. Primers used in amplification of the fragments of the four genetic markers

Primer sequence Gene marker

Gene marker Primer sequence

Size of the product (bp)

ggt TTTTAGCCATATCCGCTGCT AGCTGCTGGAGTACCAA 339

dmsA GATAGGGCATTGCGATGAGT CTTGCTAGCCCAATCAGGAG 238

Cj1585c TGTTGTGGGTTTGCTGGATA TTGCTTCACTGCATTCATCC 202

CJJ81176-1367/1371 TGCAAAGCAGGGCTAAGAAT TTATGGAGCTGGGGTGTTTC 318

4.7 Determination of antimicrobial susceptibility of Campylobacter jejuni isolates (I)

The minimum inhibitory concentrations (MICs) of ampicillin, enrofloxacin, erythromycin, gentamicin, nalidixic acid, and oxytetracycline for C. jejuni isolates from rectal faecal samples (I) were determined using a commercial broth microdilution method, VetMIC Camp (National Veterinary Institute, Uppsala, Sweden; www.sva.se/en/Target-navigation/Services--Products/VetMIC/). Epidemiological cut-off values for resistance, based on MIC distributions, were used in the interpretation of the results. A C. jejuni isolate was considered resistant to a specific antimicrobial when its MIC was distinctly higher than those of inherently susceptible C.jejuni isolates.

4.8 Statistical methods

Statistical analysis was performed using Excel or SPSS software. The �2 test was used to investigate the association between the month of sampling and the prevalence of Campylobacter spp., C. jejuni and C. hyointestinalis ssp. hyointestinalis in study I, to test the similarity in the frequencies of marker genes among the isolates from different hosts in study III, and to investigate the association between human C. jejuni genotypes and different animal reservoirs, as well as the similarity of human C. jejuni genotypes and those isolated from beef and dairy cattle herds in study IV. In addition, the host association of the combined set of the four genetic markers in study III was examined using the paired two-tailed Student’s t test.

36

5 Results

5.1 Prevalence of Campylobacter spp. in cattle at slaughter and on three dairy farms (I, II)

Campylobacter spp. were isolated from 296 of 952 (31.1%) bovine rectal faecal samples and from 33 of 948 (3.5%) bovine carcass surface samples at slaughter (Table 6). The sampled animals originated from 747 farms. The prevalence of Campylobacter spp. was higher in beef cattle than in dairy cattle in terms of the individual animals and the proportions of their farms of origin (Table 7). Among the three dairy cattle herds in study II, Campylobacter spp. were isolated from 65% (221/340) of all the faecal samples, and from one of the sponge swab samples from the drinking troughs. No campylobacters were detected in the milk samples, whereas Arcobacter butzleri was detected in three milk samples from herd 3 and in one milk sample from herd 1.

Table 6. Prevalence of Campylobacter species in bovine faecal and carcass samples at slaughter.

Faecal samples (n= 952) Carcass samples (n=948) Species

Number Prevalence Number Prevalence

Campylobacterjejuni

186 19.5 29 3.1

Campylobactercoli

21 2.2 2 0.2

Campylobacterhyointestinalis

103 10.8 2 0.2

Campylobacter spp., total 296 31.1 33 3.5

C. jejuni was the most commonly isolated thermophilic Campylobacter species in both studies (I and II). The prevalence of C. jejuni at slaughter was 19.5% (186/952). This species was more common in cattle under three years of age than in those from three to seven years of age (Table 8). In the farm study (II), C. jejuni was detected in 49.7% (169/340) of the faecal samples, and was also present in one of the drinking-trough samples. C. coli was detected in 3.2% (11/340) of the faecal samples taken on farms, and was also a minor species in samples taken at slaughter (Table 6). In herd 1, where the same ten animals were sampled on every sampling occasion, C.jejuni was isolated from all the samples of one animal, whereas two other animals tested campylobacter-negative on all occasions throughout the sampling period.

37

Table 7. Distribution of campylobacter-positive animals among beef and dairy cattle

Herd type No. of animals

No. of positive animals

Proportion of positive animals, %

No. of farms

No. of positive farms

Proportion of positive. farms, %

Beef 337 154 45.7 283 121 42.7

Dairy 615 142 23.1 463 133 28.7

Total 952 296 31.1 746 254 34.0

Beside thermophilic Campylobacter spp., C. hyointestinalis subsp.hyointestinalis was detected in bovine faeces and carcasses at slaughter (Table 6), and on average in 15.3% (52/340) of the faecal samples of the three dairy herds in study II. In addition, catalase- and urease-negative, H2S-producing Campylobacter sp. was detected in the faecal samples of herd 1 throughout the sampling period.

Table 8. The prevalence of Campylobacter spp. in Finnish cattle representing different age groups

C. jejuni C. coli C. hyointestinalis Age at slaughter

Total No. of samples Positive % Positive % Positive %

1 to 3 years 667 171 25.6 13 1.9 67 10.0

3 to 7 years 238 10 4.2 7 2.9 29 12.2

The prevalence of C. jejuni in faecal samples at slaughter showed a slightly rising trend towards the end of summer to 29.2% at its peak in August 2003 (I). The association between the sampling month and the prevalence of C. jejuni was not statistically significant. Among the three dairy herds (II), the average monthly prevalence of C. jejuni was highest (64%) in November 2006, and lowest (37%) in November 2007. The prevalences between the three herds varied widely (Figure 2). In herd 3, the prevalence was consistently higher than in the other two herds (II). Enrichment was able to detect Campylobacter spp. from 273 (37.4%), and direct culture from 32 (4.4%) of the 730 faecal samples (I). In the semiquantitative detection of study II, the levels of C. jejuni in the faecal samples were generally low (Figure 3). Of the faecal samples that tested positive, 42% were detected from the enrichment of dilution 10-2 at its peak. In herd 3, C. jejuni occurred at high levels on all sampling occasions except in August 2007.

38

0

10

20

30

40

50

60

70

80

90

100

Nov 2006 Feb 2007 Apr 2007 Aug 2007 Nov 2007

Sampling time

In-h

erd

prev

alen

ce, %

Herd 1Herd 2Herd 3

Figure 2. Prevalence of Campylobacter jejuni in three Finnish dairy cattle herds on different sampling occasions between November 2006 and November 2007.

0

10

20

30

40

50

60

70

-2 or lower -3 -4 -5 -6

Dilution of sample in the enrichment, log (g/ml)

Prop

ortio

n of

pos

itive

sam

ples

, %

Herd 1Herd 2Herd 3

Figure 3. The distribution of campylobacter levels among positive faecal samples of three dairy cattle herds determined by semiquantitative detection.

39

5.1.2 Subtypes of Campylobacter jejuni (I, II, IV)

The faecal C. jejuni isolates from study I were classified into 17 serotypes according to the 25 commercially available antisera used for typing. Isolates from 22 samples (12.5%) were untypeable. The predominant serotypes isolated from the faecal samples were Pen2 and Pen4-complex, present in 52% (96/186) of the campylobacter-positive faecal samples and in 76% (16/21) of the carcass samples. The C. jejuni isolates representing the serotype Pen2 were further divided into 23 PFGE types with SmaI, and 13 SmaI subtypes were identified among isolates representing Pen4-complex. Pen2/S1 was the most common combined sero-PFGE type among the isolates from the faecal and carcass samples (Table 9).

Table 9. The predominant combined sero/PFGE types of Campylobacter jejuni isolates from bovine faecal samples at slaughter

Penner

serotype SmaI subtype No. of positive samples

% of positive samples

2 S1 19 10.8 2 S5 10 5.7 2 S11 7 4.0 2 S18 5 2.8

4-complex S2 10 5.7 4-complex S3 7 4.0

12 S7 8 4.5

1,44 S6 5 2.8

Total 71 40.3

In total, PFGE with SmaI restriction enzyme identified 56 different subtypes of C. jejuni among the 330 isolates from the faecal samples of slaughter cattle and 20 subtypes among the 33 isolates from the carcass samples taken before chilling (I). Isolates from 30 C. jejuni-positive animals (16.1%) and from 11 (33.3%) carcasses represented unique subtypes. The DNA from five faecal isolates was not digestible with SmaI. In study II, a total of thirteen SmaI genotypes were identified among the C. jejuni isolates (n=403) from the three dairy herds. One to four SmaI subtypes were detected from each of the herds on each sampling occasion, except in August 2007, when no campylobacters were isolated from herd 2. In herds 1 and 3, however, two subtypes

40

persisted throughout the entire sampling period from November 2006 to November 2007. A few additional types emerged in August 2007, whereas in herd 2, only two C. jejuni subtypes occurred during the entire sampling period, and no Campylobacter spp. were detected in August 2007. In study I, C. jejuni isolates from animals originating from the same farm during the same sampling (16 occasions) represented indistinguishable or related SmaI types on nine occasions and unrelated types on five occasions. C. jejuni isolates from the same farms on two sampling occasions (six farms) represented unrelated subtypes. Two different SmaI subtypes were detected in 3 of the 169 positive faecal samples in study II, and in study I, the PFGE of multiple isolates from 106 faecal samples identified different SmaI types in eight samples. Two different types were detected from two carcass samples. On 12 sampling occasions, the faecal and carcass samples from the same animal yielded indistinguishable C. jejuni SmaI types (I). In addition, identical subtypes were isolated from one animal’s faecal sample and from another animal’s carcass sample during six samplings. In study II, isolates from each of the animals in herd 1 that yielded multiple campylobacter-positive samples were consistently indistinguishable, with the exception of a previous carrier of subtype S7, from which subtype S64 was isolated after two negative samples. Furthermore, the C. jejuni isolates from the drinking trough at farm 3 represented the most frequently detected two SmaI subtypes among animals in that herd. In study IV, PFGE with SmaI restriction identified 43 subtypes among the 175 C. jejuni isolates from human domestic infections between June and August 2003, and 15 subtypes among the 43 isolates from chicken slaughter batches between May and August 2003. SmaI was unable to type 18 isolates from humans and one from chickens. Bovine faecal isolates from the entire year (n=186) and carcass isolates from May to August 2003 (n=15) represented a total of 61 subtypes. Fourteen SmaI subtypes of C. jejuni (32.6% of all 43 human subtypes) representing 114 (65.1%) of 175 human isolates overlapped with those of chicken or bovine isolates. In total, 83.7% (36/43) of chicken isolates and 30.8% (62/201) of bovine isolates represented SmaI subtypes shared with humans. Further subtyping of 212 C. jejuni isolates (114 human, 36 chicken, and 62 cattle isolates), representing the 14 overlapping SmaI subtypes with KpnI restriction enzyme yielded 44 subtypes, 17 of which were shared between human and animal isolates (Table 10). The combined type S6/K12 predominated among the isolates from human patients (12%), and occurred in chickens and cattle as well. In total, the SmaI/KpnI profiles of 97 (55.4%) human isolates were indistinguishable from those of chicken or cattle isolates. The overlapping combined SmaI/KpnI subtypes accounted for 69.8% (30/43) of the chicken isolates and 15.9% (32/201) of the cattle isolates. The occurrence of identical SmaI/KpnI subtypes with human C. jejuni isolates was significantly associated with animal host species (P < 0.001).

41

All ten bovine subtypes overlapping with those of humans represented isolates from dairy cattle (n=31), with the exception of S22/K16, isolated from only one beef cattle. The occurrence of identical SmaI/KpnI subtypes with human C. jejuni isolates in cattle was not significantly related to herd type (P =0.056). A temporal association of the SmaI/KpnI subtypes among isolates from chickens and patients was possible in 55 (31.4%) of 175 human infections (Table 11). Isolates from 27 (15.4%) of human cases with no temporal relation to chickens were identical to bovine isolates.

Table 10. Occurrence of overlapping SmaI/KpnI subtypes of Campylobacter jejuni in domestically acquired human sporadic infections, chickens and cattle in Finland in summer 2003

Origin of isolates

PFGE subtype Human Chicken Cattle

SmaI KpnI No. of isolates

% of isolates

No. of isolates

% of isolates

No. of isolates

% of isolates

S4 K29 1 0.6 1 2.3 1 0.5 S5 K27 1 0.6 0 0.0 10 4.9 S6 K12 21 12.0 2 4.7 7 3.4 S7 K1 12 6.9 2 4.7 7 3.4 S7 K2 4 2.3 2 4.7 2 1.0 S7 K3 17 9.7 2 4.7 1 0.5

S22 K16 1 0.6 0 0.0 1 0.5 S54 K10 6 3.4 2 4.7 0 0.0 S54 K11 3 1.7 1 2.3 0 0.0 S64 K19 7 4.0 1 2.3 1 0.5 S66 K18 4 2.3 0 0.0 1 0.5 S74 K4 5 2.9 8 18.6 0 0.0 S74 K5 8 4.6 4 9.3 1 0.5 S74 K7 2 1.1 2 4.7 0 0.0 S76 K20 3 1.7 1 2.3 0 0.0 S77 K30 1 0.6 1 2.3 0 0.0 S78 K6 1 0.6 1 0.0 0 0.0

Isolates of shared subtypes 97 55.4 30 69.8 32 15.6

Total No. of isolates 175 43 201

Tab

le 1

1. T

empo

ral a

ssoc

iatio

n be

twee

n C

ampy

loba

cter

jeju

ni is

olat

es fr

om h

uman

s an

d ch

icke

n sl

augh

ter

batc

hes

in s

umm

er 2

003

in F

inla

nd

SmaI

/Kpn

I N

umbe

r of h

uman

isol

ates

June

Ju

ly

Aug

ust

Tota

l Su

btyp

e as

soci

ated

no

t ass

ocia

ted

asso

ciat

ed

not a

ssoc

iate

d as

soci

ated

no

t ass

ocia

ted

asso

ciat

ed

not a

ssoc

iate

d S4

/K29

0

0 0

1 0

0 0

1 S6

/K12

0

0 7

0 14

0

21

0 S7

/K1

0 1

8 0

3 0

11

1 S7

/K2

0 0

0 4

0 0

0 4

S7/K

3 0

2 1

5 9

0 10

7

S54/

K10

0

0 0

6 0

0 0

6 S5

4/K

11

0 0

0 1

1 1

1 2

S64/

K19

0

0 0

5 1

1 1

6 S7

4/K

4 0

0 5

0 0

0 5

5 S7

4/K

5 0

0 0

8 0

0 0

8 S7

4/K

7 0

0 0

0 2

0 2

2 S7

6/K

20

0 0

0 0

3 0

3 3

S77/

K30

0

0 0

0 1

0 1

1 S7

8/K

6 0

0 0

1 0

0 0

1 To

tal

0 3

21

31

34

2 55

47

To

tal N

o. o

f hu

man

isol

ates

11

10

6

58

17

5

42

43

5.1.3 Occurrence of genetic markers among Campylobacter jejuni isolates from humans, chickens and cattle (III)

The �-glutamyl transpeptidase and dmsA genes were more frequently detected among human and chicken C. jejuni isolates than among bovine isolates. In addition, dmsA-positive chicken isolates occurred with a similar high annual frequency in 2003, 2006, and 2007. In contrast, the Cj1585 oxidoreductase and the CJJ81176-1371 serine protease genes were more common among the bovine isolates than among the human and chicken isolates (Table 12). The bovine isolates differed significantly (P <0.05) from human and chicken isolates in the t test.

Table 12. Occurrence of four marker genes (ggt, dmsA, Cj1585c and CJJ81176-1371) in Campylobacter jejuni isolates from humans, chickens and cattle

Number of isolates harbouring the gene (%)

Marker gene Human (n=309)

Chicken (n=205)

Cattle (n=131)

ggt 169 (54.7) 75 (36.6) 11 (8.4)

dmsA 256 (82.8) 151 (73.3) 18 (13.7)

Cj1585c 99 (32.0) 49 (23.9) 83 (62.6)

CJJ81176-1367/1371 117 (37.8) 74 (36.1) 96 (73.3)

5.1.4 Antimicrobial susceptibility of bovine Campylobacterjejuni isolates (I)

Of the 187 C. jejuni isolates examined for antimicrobial susceptibility, 16 (9%) proved resistant to at least one of the antimicrobials tested (Table 13). Resistance to nalidixic acid was most common. Six of the 11 nalidixic acid-resistant isolates were also resistant to enrofloxacin. None of the isolates presented multiresistance.

Tab

le 1

3. D

istr

ibut

ion

of m

inim

um in

hibi

tory

con

cent

ratio

ns (

MIC

s) a

mon

g bo

vine

Cam

pylo

bact

er je

juni

isol

ates

(n=

187)

D

istri

butio

n of

MIC

s (m

g/1)

b

Su

bsta

nce

% re

sist

ant i

sola

tes

(95%

CIa )

� 0.

030.

06

0.12

0.

25

0.5

1 2

4 8

16

32

64

128

256

512

>512

Am

pici

llin

1.6

(0.3

-4.6

)

5.

9 7.

0 41

.2

40.1

3.

2 1.

1 0.

5 1.

1

Enro

floxa

cin

3.2

(1.2

-6.9

) 1.

1 8.

0 49

.2

33.7

4.

8 1.

1 0.

0 1.

6 0.

5c

Eryt

hrom

ycin

0

(0.0

-2.0

)

1.

1 1.

6 22

.5

51.9

20

.9

2.1

Gen

tam

icin

0

(0.0

-2.0

)

3.2

54.0

42

.2

0.5

1.1

Nal

idix

ic a

cid

5.9

(3.0

-10.

3)

1.6

15.5

61

.0

16.0

3.7

0.0

0.5

1.6

Tetra

cycl

ine

1.1

(0.1

-3.8

)

92.0

5.

9 0.

5 0.

5 0.

5 0.

5

Bol

d ve

rtica

l lin

es i

ndic

ate

cut-o

ff v

alue

s fo

r re

sist

ance

. H

atch

ed f

ield

s de

note

ran

ge o

f di

lutio

ns t

este

d fo

r ea

ch

subs

tanc

e.

a CI,

conf

iden

ce in

terv

al.

b MIC

s equ

al to

or l

ower

than

the

low

est c

once

ntra

tion

test

ed a

re g

iven

as t

he lo

wes

t con

cent

ratio

n.

c MIC

gre

ater

than

the

high

est c

once

ntra

tion

in th

e ra

nge

of d

ilutio

ns te

sted

.

44

6 Discussion

6.1 Campylobacter spp. in Finnish cattle

The sampling in our survey on Campylobacter spp. in cattle at slaughter (I) covered all major Finnish slaughterhouses representing 98% of the cattle slaughtered in Finland in 2003. The prevalence ofCampylobacter spp. among Finnish cattle was lower than that in several other studies (Garcia et al. 1985, Stanley et al. 1998, Beach et al. 2002, Johnsen et al. 2006, Milnes et al. 2008). However, the results from different studies are not fully comparable due to different study designs and laboratory methods. The enrichment in our survey detected 7.5 times more campylobacter-positive faecal samples than did direct plating. The high number of false negative results obtained from direct plating probably reflected the low levels of Campylobacter spp. in the faeces of slaughter cattle, which is actually consistent with study II. In this study, which focused on three dairy herds on farms, and in accordance with previous studies (Stanley et al. 1998, Nielsen 2002, Heuvelink et al. 2009), most of the animals excreted lower levels of Campylobacter spp. than levels of C. jejuni in the caeca of chickens, which can reach up to 108 cfu/g (Reich et al. 2008). Only a few animals in study II excreted high numbers (>106 cfu/g) as determined by the semiquantitative detection method, which was based on the enrichment of 10-fold dilutions of the faecal samples (NCFA [Nordic Committee on Food Analysis] 2007). The higher prevalence of Campylobacter spp. observed among beef cattle (I) may reflect the age distribution of the animals, because most of the beef cattle at slaughter were under three years of age, and the overall prevalence in that age group was higher than among older animals. Similar observations of the age-related prevalence of campylobacters in cattle are common in previous studies as well (Giacoboni et al. 1993, Nielsen 2002, Johnsen et al. 2006, Gilpin et al. 2008b). However, some studies have suggested that the higher prevalence of Campylobacter spp. in beef cattle could derive from different farming practices, such as different feed and higher animal density than among dairy cattle (Wesley et al. 2000, Minihan et al. 2004). Similar to our results from studies I and II, most other studies on bovine campylobacters have reported C. jejuni as the most common Campylobacter sp. in cattle (Wesley et al. 2000, Berry et al. 2006, Madden et al. 2007, Milnes et al. 2008, Ragimbeau et al. 2008), whereas others, applying specific methods, have detected the predominance of other species such as C. hyointestinalis ssp. hyointestinalis or C. lanienae (Grau 1988, Atabay and Corry 1998, Inglis et al. 2003, Pezzotti et al. 2003). We detected C. hyointestinalis

45

in 10.8% of the faecal samples at slaughter, but no C. lanienae, which is anaerobic and requires specific cultivation conditions. However, on the basis of current knowledge these two species appear to be minor human pathogens (Lastovica and Allos 2008), whereas C. jejuni is the most commonly reported species in human infections (Baker et al. 2007, EFSA 2010b). The results from study II indicate that dairy cattle can be long-term carriers of C. jejuni with varying shedding patterns among herds. Unlike in some other studies (Stanley et al. 1998, Kwan et al. 2008b, Grove-White et al. 2010), we cannot draw general conclusions in regard to seasonal variation of shedding due to the small number of herds, the few sampling occasions and the study period of only one year. While in study I the prevalence of C. jejuni peaked in August, study II detected no peak in the prevalence of C. jejuni among the herds in August, although the herds had been grazing since May, and were therefore probably exposed to a variety of potential environmental sources of campylobacters (Oporto et al. 2007, Grove-White et al. 2010). None of the herds had access to natural water sources, however, which may indirectly illustrate the importance of natural waters as a reservoir of campylobacters for cattle during grazing (Humphrey and Beckett 1987, Hänninen et al. 1998). The last sampling that occurred after grazing, however, yielded high prevalences in all of the herds, possibly due to changes in the diet (Stanley et al. 1998). In addition, the prevalence of C. jejuni rose in herd 3 during indoor housing in winter. The water trough samples taken on the last sampling occasion provide a plausible explanation: the predominating C. jejuni subtype in herd 3 was present at a detectable level in one of those samples, and probably contributed to the persistent colonisation of the herd when housed indoors (Minihan et al. 2004, Ellis-Iversen et al. 2009a). Unfortunately, we took no samples from the dug well, which was the drinking water supply for herd 3. The persons living on the farm, however, consumed water obtained from the same supply without any symptoms of the disease.

6.2 The diversity of Campylobacter jejuni in Finnish cattle

The C. jejuni sero/PFGE types in bovine faecal samples revealed high diversity in the slaughterhouse survey (I). Nevertheless, in studies I and II only one subtype was usually detected in the samples of individual animals, from which up to six isolates were genotyped. In addition, C. jejuni isolates from different animals originating from the same farm in study I consistently represented identical subtypes on the same sampling occasion. Moreover, in the three cattle herds in study II, only one or two persistent PFGE subtypes of C. jejuni were detected among each herd throughout the study, although earlier studies have reported a wider range of subtypes in adult cattle (Nielsen 2002, Kwan et al. 2008b). The presence of a small number of subtypes suggests only a few sources of C. jejuni or re-infection with the same

46

strains during the study period (Nielsen 2002, Minihan et al. 2004). In two of the herds, additional subtypes of C. jejuni occurred mainly during the grazing period, thus indicating new sources from the environment (Brown et al. 2004, Oporto et al. 2007, Grove-White et al. 2010). Beside the few sources on the farms, the presence of only a few subtypes of C. jejuni in herds may suggest ecological competition between strains in bovine intestines (Kwan et al. 2008b). Subtypes available at an early stage of an animal’s life, probably have fewer competitors in the immature gut, whereas later exposure to other subtypes may result in only intermittent shedding due to the competitive advantage of the earlier colonisers. In addition, re-infection with the same few subtypes present in a herd is probably an important contributor to the colonisation of animals, as was apparent in herd 3 in study II (Ellis-Iversen et al. 2009b) Subtyping of the C. jejuni isolates from the same animals on different sampling occasions in herd 1 revealed that some of the animals were intermittent carriers of campylobacters, whereas others appeared to be persistent shedders of a single subtype (Hänninen et al. 1998, Gilpin et al. 2008b, Kwan et al. 2008b). In addition, one of the animals was campylobacter-negative in all samplings, which may indicate acquired immunity, different intestinal microbiota or other individual characteristics that prevent colonisation (Minihan et al. 2004).

6.3 Chickens and cattle as sources of Campylobacterjejuni in sporadic human infections in Finland

6.3.1 Comparison of subtypes of Campylobacter jejuni from human infections, chickens and cattle

Serotyping, while comparable between laboratories, offers low discriminatory power in the typing of C. jejuni. Consequently, serotyping results are merely suggestive, and inconclusive for source attribution. The predominant serotypes of C. jejuni identified among cattle (I) - Pen2, Pen4-complex, Pen1,44 and Pen12 - occur in domestic human infections in Finland as well (Rautelin and Hänninen 1999, Vierikko et al. 2004, Nakari et al. 2005, Schönberg-Norio et al. 2006). Studies from other countries have also reported the common presence of Pen2 and Pen4-complex in the faeces of dairy cattle (Nielsen et al. 1997, Nielsen 2002, Devane et al. 2005, Ishihara et al. 2006), which may indicate the adaptation of these serotypes of C.jejuni to the bovine intestinal tract. In addition, the serotype Pen2 was present only in human isolates representing rural areas of Finland in a previous study that compared different geographical areas (Schönberg-Norio et al. 2006), and is uncommon in Finnish chickens (Perko-Mäkelä et al. 2002), which may indicate, in accordance with studies from other countries (Studahl and Andersson 2000, Baker et al. 2007, Garrett et al. 2007, Strachan et al. 2009) the contribution of cattle as source of C. jejuni in human infections in rural areas of Finland.

47

The comparison of domestic human, chicken and bovine isolates of C.jejuni focused on the isolates present during the summer months from June to August 2003, because the incidence of human campylobacteriosis in Finland consistently peaks in July-August. Furthermore, most of the human infections are domestically acquired in summer, whereas those in winter are mainly travel-related (National Institute for Health and Welfare 2009). Our study included domestic human isolates from nine clinical microbiology laboratories across the country, chicken strains from two Finnish slaughterhouses representing approximately 80% of the total slaughter volume during the study period, all bovine faecal strains isolated at slaughter between January and December 2003 (assuming that the shedding of the subtypes detected at slaughter was similar to that in study II and continued in the herds throughout the year), and all isolates from bovine carcasses during the summer of 2003. Due to the relatively short time-frame of the study in a geographically defined area, we considered PFGE with two restriction enzymes suitable for comparison of the isolates as a highly discriminating subtyping method. As with the bovine isolates, high genotypic diversity was apparent among the human C. jejuni isolates, whereas the number of different subtypes from chickens was small due to the low prevalence of C.jejuni in Finnish chicken slaughter batches (Perko-Mäkelä et al. 2002, EFSA 2010c). Only one isolate per chicken slaughter batch was available for comparison. However, isolation of more than one strain from each campylobacter-positive batch would probably not have affected the outcome, because in the majority of Finnish campylobacter-positive chicken flocks, only one C. jejuni subtype is present in each growing batch (Hakkinen and Kaukonen 2009). Isolates representing genotypes indistinguishable from those of chickens or cattle were present in 55.4% of the human infections. Considering the temporal association of chicken isolates, 31.4 % of the human cases could have originated from chickens, similar to the previous estimate from the summer of 1999 (Kärenlampi et al. 2003). The remaining temporally unrelated subtypes that were identical to those from cattle represented 15.4% of the human infections. In addition, subtypes shared only between humans and cattle were present in 3.4% of the human cases. The total proportion of human domestic infections of bovine origin during the summer 2003 in Finland could thus have been approximately 19 %. A previous Finnish MLST study observed a high degree of overlap (61%) between human and chicken isolates, whereas overlap was very low (5.7%) between human and bovine isolates (Kärenlampi et al. 2007). The number of bovine isolates was low in the study, however, and the collections of human isolates represented a different geographical area, and thus probably different sources of infection as well (Schönberg-Norio et al. 2006). In contrast to the study of (Kärenlampi et al.) (2007), which analysed human isolates from a more urban area in southern Finland, our isolates represented the entire country and covered rural areas more extensively. Recent research elsewhere has focused increasingly

48

on the different exposures among populations in urban and rural areas and has identified, for example, increasing ruminant density and contact with cattle as risk factors (Studahl and Andersson 2000, Kapperud et al. 2003, Nygård et al. 2004). With one exception, subtypes shared between human and cattle originated from dairy cattle, although C. jejuni was more common in beef cattle herds (I). Moreover, none of the C. jejuni subtypes isolated from carcasses was present among human isolates, thus supporting the conclusion of the prevalence study (I), which suggests that beef is of minor importance as source of campylobacters in human infections due to the low prevalence of Campylobacter spp. on carcasses. Because air-chilling further reduces the contamination of carcasses with campylobacters due to the sensitivity of these organisms to oxygen and drying (Oosterom et al. 1983, Grau 1988), the survival of campylobacters on retail beef is unlikely. Milk, instead, can permit longer survival of Campylobacter spp., if failures in milking hygiene lead to faecal contamination with these organisms (Doyle and Roman 1982). Despite the high prevalence of C. jejuni in the dairy herds, no Campylobacter spp. occurred in the milk samples in study II, which indicates adequate milking hygiene on the participating farms. Milkborne outbreaks are rare in Finland, because up to 97% of milk is delivered to dairies, and the consumption of unpasteurised milk is uncommon (http://www.maataloustilastot.fi/en/node/540). The food-related transmission of bovine Campylobacter spp. to humans therefore appears insignificant, whereas occupational and environmental routes require further consideration. In particular, the presence of human pathogenic campylobacters among dairy herds is of concern because of the long life-span of dairy cattle, during which persistent carriers of campylobacters in the herds increase the environmental load of these organisms in rural areas.

6.3.2 Genetic markers in differentiation of the sources of Campylobacter jejuni in human infections

Genetic markers revealed higher similarity among human and chicken C. jejuni isolates than among human and bovine isolates. The controversy of the PFGE result may partially stem from the different time frames of studies III and IV, and the different geographical origin of human isolates, which were obtained from more urban areas in southern Finland in study III than in study IV. The controversial results may therefore reflect differences in rural and urban exposures (Studahl and Andersson 2000, Schönberg-Norio et al. 2006, Garrett et al. 2007). Nevertheless, the results also indicate differences in the metabolic characteristics of C. jejuni strains isolated from chicken and cattle, which supports the previously observed host adaptation of C. jejuni (Dingle et al. 2001, Champion et al. 2005, McCarthy et al. 2007). In our study III, the ggt gene, which previously seemed to relate to the prolonged intestinal colonisation of C. jejuni in chickens (Barnes et al.

49

2007) and to the enhanced colonisation of human intestinal tissues due to the acquired ability of C. jejuni to utilise glutathione and glutamine as sources of amino acids (Hofreuter et al. 2008), was more common among chicken and human isolates than among bovine C. jejuni isolates. Similarly, the subunit of the putative anaerobic DMSO oxidoreductase gene, dmsA, was rare among bovine isolates, but occurred frequently among human and chicken isolates in our study. In a previous study, C. jejuni colonisation in chickens was associated with the presence of this oxidoreductase (Hiett et al. 2008), which may contribute to the virulence of C. jejuni also (Hofreuter et al. 2006). Another putative oxidoreductase gene, Cj1585, was more common in bovine C. jejuni isolates, which may indicate that the Cj1585 type oxidoreductase system is preferential in the oxygen-restricted environment of the bovine intestine. In addition, the bovine isolates were more frequent carriers of the subtilase-type serine protease gene CJJ81176-1367/1371. The presence of the serine protease Cj1371 apparently relates to the tolerance of oxidative stress in C. jejuni (Garenaux et al. 2008), but its contribution to the pathogenesis of C.jejuni is unknown. In several other pathogens, such as Vibrio cholerae, Shigella dysenteriae and some VTEC strains, the production of subtilase cytotoxins, which harbour a subunit homologous with subtilase-like serine proteases, appears to be important to virulence (Beddoe et al. 2010). For example, a highly cytotoxic subtilase toxin (SubAB) of some VTEC strains causes in mice lesions that resemble those in patients with HUS (Paton et al. 2004, Wang et al. 2007).

6.4 Antimicrobial susceptibility of bovine Campylobacter jejuni isolates

The low prevalence of resistance to antimicrobials among bovine C.jejuni isolates is probably a consequence of the prudent veterinarian use of these agents in Finland. We applied the epidemiological cut-off values in the determination of susceptibility, according to the recommendation of EFSA for monitoring purposes. Using the same values, the resistance levels of bovine C. jejuni isolates in seven European countries during the period from 2004 to 2007 were substantially higher: the average tetracycline resistance varied between 23% and 33% and nalidixic acid resistance from 23% and 35% (EFSA 2010a). The resistance levels among domestic human Campylobacter isolates and chicken isolates have also been low in Finland, and resistant strains occur mainly in travel-related infections (Rautelin et al. 2003, Feodoroff et al. 2009, EFSA 2010a).

50

7 Conclusions

1. Finnish cattle appeared to be a constant reservoir of C.jejuni, the most common Campylobacter species in human infections. The level of faecal excretion of C. jejuni was usually low, so enrichment is essential for optimal isolation of the organism from bovine faecal samples.

2. Beef cattle appeared to be more frequent carriers of C.

jejuni than were dairy cattle. The contamination of carcasses was low at slaughter, however, and the isolates from carcasses represented different PFGE types from those in humans. Beef therefore appears to be an insignificant source of campylobacters in human infections.

3. The resistance to antimicrobials was low among bovine C.

jejuni isolates, and no multiresistance occurred. This is probably due to the prudent use of antimicrobials in Finnish animal production, and indicates a low risk for human infections by resistant strains of bovine origin.

4. Diverse shedding patterns of C. jejuni occurred among

both dairy cattle herds and individual animals. The same few subtypes of C. jejuni were able to persist in a dairy herd for more than one year. Ecological competition in the colonisation of the bovine intestinal tract may occur between different subtypes of C. jejuni. In addition, individual animals can be resistant to colonisation. The faecal contamination of water troughs can maintain colonisation in cattle herds during indoor housing. At pasture, however, preventing access to natural waters can limit colonisation. Despite the high percentage of animals in dairy herds shedding Campylobacter spp. in their faeces, adequate milking hygiene could prevent the contamination of milk.

5. The distribution of chicken and bovine isolates based on

the presence of genetic markers supported the previous observations of the host adaptation of C. jejuni strains apparently as a response to different type of oxidative stress and metabolic demands in the intestinal tracts of these animal species. In addition, differences in the genetic markers may suggest differences in the virulence of C. jejuni strains from chickens and cattle.

6. The isolates from 55.4% of sporadic domestic human

infections during the seasonal peak in 2003 represented identical PFGE subtypes with C. jejuni isolates from

51

chickens and cattle, especially dairy cattle. The proportion of human cases temporally associated with chicken isolates was 31.1%, and approximately 19% of human infections were possibly related to cattle, suggesting an important role for Finnish cattle, besides chickens, as a source of C. jejuni in human infections, although common sources of C. jejuni in humans, chickens and cattle are also possible. Our results suggest that food is probably a minor route of transmission of bovine C. jejuni, and the sources of C. jejuni in human infections in rural areas may differ from those in urban areas in Finland.

52

References

Acik, M.N. and Cetinkaya, B. (2005) The heterogeneity of Campylobacter jejuni and Campylobacter coli strains isolated from healthy cattle. Lett Appl Microbiol 41, 397-403.

Adak, G.K., Meakins, S.M., Yip, H., Lopman, B.A. and O'Brien, S.J. (2005) Disease risks from foods, England and Wales, 1996-2000. Emerg Infect Dis 11, 365-372.

Anderson, S.A., Yeaton Woo, R.W. and Crawford, L.M. (2001) Risk assessment of the impact on human health of resistant Campylobacterjejuni from fluoroquinolone use in beef cattle. Food Control, 13-25.

Atabay, H.I. and Corry, J.E. (1998) The isolation and prevalence of campylobacters from dairy cattle using a variety of methods. J Appl Microbiol 84, 733-740.

Bachmann, H.P. and Spahr, U. (1995) The fate of potentially pathogenic bacteria in Swiss hard and semihard cheeses made from raw milk. J Dairy Sci 78, 476-483.

Bae, W., Kaya, K.N., Hancock, D.D., Call, D.R., Park, Y.H. and Besser, T.E. (2005) Prevalence and antimicrobial resistance of thermophilic Campylobacter spp. from cattle farms in Washington State. Appl Environ Microbiol 71, 169-174.

Baker, M., Wilson, N., Ikram, R., Chambers, S., Shoemack, P. and Cook, G. (2006) Regulation of chicken contamination is urgently needed to control New Zealand's serious campylobacteriosis epidemic. N Z Med J 119, U2264.

Baker, M.G., Sneyd, E. and Wilson, N.A. (2007) Is the major increase in notified campylobacteriosis in New Zealand real? Epidemiol Infect135, 163-170.

Barnes, I.H., Bagnall, M.C., Browning, D.D., Thompson, S.A., Manning, G. and Newell, D.G. (2007) Gamma-glutamyl transpeptidase has a role in the persistent colonization of the avian gut by Campylobacter jejuni. Microb Pathog 43, 198-207.

Beach, J.C., Murano, E.A. and Acuff, G.R. (2002) Prevalence of Salmonella and Campylobacter in beef cattle from transport to slaughter. J Food Prot 65, 1687-1693.

Beddoe, T., Paton, A.W., Le Nours, J., Rossjohn, J. and Paton, J.C. (2010) Structure, biological functions and applications of the AB(5) toxins. Trends Biochem Sci. 35, 411-418.

53

Berry, E.D., Wells, J.E., Archibeque, S.L., Ferrell, C.L., Freetly, H.C. and Miller, D.N. (2006) Influence of genotype and diet on steer performance, manure odor, and carriage of pathogenic and other fecal bacteria. II. Pathogenic and other fecal bacteria. J Anim Sci 84, 2523-2532.

Besser, T.E., Lejeune, J.T., Rice, D.H., Berg, J., Stilborn, R.P., Kaya, K., Bae, W. and Hancock, D.D. (2005) Increasing prevalence of Campylobacter jejuni in feedlot cattle through the feeding period. ApplEnviron Microbiol 71, 5752-5758.

Birk, T. and Knochel, S. (2009) Fate of food-associated bacteria in pork as affected by marinade, temperature, and ultrasound. J Food Prot 72, 549-555.

Black, R.E., Levine, M.M., Clements, M.L., Hughes, T.P. and Blaser, M.J. (1988) Experimental Campylobacter jejuni infection in humans. JInfect Dis 157, 472-479.

Blaser, M. and Engberg, J. (2008) Clinical aspects of Campylobacter jejuni and Campylobacter coli infections. In ed. Nachamkin, I., Szymanski, C.M. and Blaser, M.J. pp. 99-121. Washington, DC, USA: ASM Press.

Boes, J., Nersting, L., Nielsen, E.M., Kranker, S., Enoe, C., Wachmann, H.C. and Baggesen, D.L. (2005) Prevalence and diversity of Campylobacter jejuni in pig herds on farms with and without cattle or poultry. J Food Prot 68, 722-727.

Bostan, K., Aydin, A. and Ang, M.K. (2009) Prevalence and antibiotic susceptibility of thermophilic Campylobacter species on beef, mutton, and chicken carcasses in Istanbul, Turkey. Microb Drug Resist 15, 143-149.

Broman, T., Palmgren, H., Bergstrom, S., Sellin, M., Waldenstrom, J., Danielsson-Tham, M.L. and Olsen, B. (2002) Campylobacter jejuni in black-headed gulls (Larus ridibundus): prevalence, genotypes, and influence on C. jejuni epidemiology. J Clin Microbiol 40, 4594-4602.

Broman, T., Waldenstrom, J., Dahlgren, D., Carlsson, I., Eliasson, I. and Olsen, B. (2004) Diversities and similarities in PFGE profiles of Campylobacter jejuni isolated from migrating birds and humans. JAppl Microbiol 96, 834-843.

Brown, P.E., Christensen, O.F., Clough, H.E., Diggle, P.J., Hart, C.A., Hazel, S., Kemp, R., Leatherbarrow, A.J., Moore, A., Sutherst, J., Turner, J., Williams, N.J., Wright, E.J. and French, N.P. (2004) Frequency and spatial distribution of environmental Campylobacter spp. Appl Environ Microbiol 70, 6501-6511.

54

Busato, A., Hofer, D., Lentze, T., Gaillard, C. and Burnens, A. (1999) Prevalence and infection risks of zoonotic enteropathogenic bacteria in Swiss cow-calf farms. Vet Microbiol 69, 251-263.

Butzler, J.P., Dekeyser, P., Detrain, M. and Dehaen, F. (1973) Related vibrio in stools. J Pediatr 82, 493-495.

Bywater, R., Deluyker, H., Deroover, E., de Jong, A., Marion, H., McConville, M., Rowan, T., Shryock, T., Shuster, D., Thomas, V., Valle, M. and Walters, J. (2004) A European survey of antimicrobial susceptibility among zoonotic and commensal bacteria isolated from food-producing animals. J Antimicrob Chemother 54, 744-754.

Carrique-Mas, J., Andersson, Y., Hjertqvist, M., Svensson, A., Torner, A. and Giesecke, J. (2005) Risk factors for domestic sporadic campylobacteriosis among young children in Sweden. Scand J Infect Dis 37, 101-110.

Centers for Disease Control and Prevention (CDC) (2009) Campylobacter jejuni infection associated with unpasteurized milk and cheese--Kansas, 2007. MMWR Morb Mortal Wkly Rep 57, 1377-1379.

Champion, O.L., Gaunt, M.W., Gundogdu, O., Elmi, A., Witney, A.A., Hinds, J., Dorrell, N. and Wren, B.W. (2005) Comparative phylogenomics of the food-borne pathogen Campylobacter jejuni reveals genetic markers predictive of infection source. Proc Natl Acad Sci U S A 102, 16043-16048.

Chatre, P., Haenni, M., Meunier, D., Botrel, M.A., Calavas, D. and Madec, J.Y. (2010) Prevalence and antimicrobial resistance of Campylobacter jejuni and Campylobacter coli isolated from cattle between 2002 and 2006 in France. J Food Prot 73, 825-831.

Chaudhuri, R.R., Loman, N.J., Snyder, L.A., Bailey, C.M., Stekel, D.J. and Pallen, M.J. (2008) xBASE2: a comprehensive resource for comparative bacterial genomics. Nucleic Acids Res 36, D543-6.

Clark, C.G., Price, L., Ahmed, R., Woodward, D.L., Melito, P.L., Rodgers, F.G., Jamieson, F., Ciebin, B., Li, A. and Ellis, A. (2003) Characterization of waterborne outbreak-associated Campylobacterjejuni, Walkerton, Ontario. Emerg Infect Dis 9, 1232-1241.

CLSI, ed. (2008) Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals; Approved Standard-Third ed. [ISBN Number 1-56238-659-X]. CLSI Document M31-A3. Wayne, PA, USA: Clinical and Laboratory Standards Institute.

CLSI (2006) Methods for Antimicrobial Dilution and Disk Susceptibility Testing of Infrequently Isolated or Fastidious Bacteria; Approved Guideline. CLSI document M45-A. Wayne, Pennsylvania, USA: Clinical and Laboratory Standards Institute.

55

Colles, F.M., Jones, K., Harding, R.M. and Maiden, M.C. (2003) Genetic diversity of Campylobacter jejuni isolates from farm animals and the farm environment. Appl Environ Microbiol 69, 7409-7413.

de Boer, P., Duim, B., Rigter, A., van Der Plas, J., Jacobs-Reitsma, W.F. and Wagenaar, J.A. (2000) Computer-assisted analysis and epidemiological value of genotyping methods for Campylobacterjejuni and Campylobacter coli. J Clin Microbiol 38, 1940-1946.

Debruyne, L., Gevers, D. and Vandamme, P. (2008) Taxonomy of the family Campylobacteraceae. In Campylobacter ed. Nachamkin, I., Szymanski, C.M. and Blaser, M. pp. 3-25. Washington D.C.: ASM Press.

Dekeyser, P., Gossuin-Detrain, M., Butzler, J.P. and Sternon, J. (1972) Acute enteritis due to related vibrio: first positive stool cultures. JInfect Dis 125, 390-392.

Denno, D.M., Keene, W.E., Hutter, C.M., Koepsell, J.K., Patnode, M., Flodin-Hursh, D., Stewart, L.K., Duchin, J.S., Rasmussen, L., Jones, R. and Tarr, P.I. (2009) Tri-county comprehensive assessment of risk factors for sporadic reportable bacterial enteric infection in children. JInfect Dis 199, 467-476.

Desai, M., Logan, J.M., Frost, J.A. and Stanley, J. (2001) Genome sequence-based fluorescent amplified fragment length polymorphism of Campylobacter jejuni, its relationship to serotyping, and its implications for epidemiological analysis. J Clin Microbiol 39, 3823-3829.

Devane, M.L., Nicol, C., Ball, A., Klena, J.D., Scholes, P., Hudson, J.A., Baker, M.G., Gilpin, B.J., Garrett, N. and Savill, M.G. (2005) The occurrence of Campylobacter subtypes in environmental reservoirs and potential transmission routes. J Appl Microbiol 98, 980-990.

Dilworth, C.R., Lior, H. and Belliveau, M.A. (1988) Campylobacter enteritis acquired from cattle. Can J Public Health 79, 60-62.

Dingle, K.E., Colles, F.M., Ure, R., Wagenaar, J.A., Duim, B., Bolton, F.J., Fox, A.J., Wareing, D.R. and Maiden, M.C. (2002) Molecular characterization of Campylobacter jejuni clones: a basis for epidemiologic investigation. Emerg Infect Dis 8, 949-955.

Dingle, K.E., Colles, F.M., Wareing, D.R., Ure, R., Fox, A.J., Bolton, F.E., Bootsma, H.J., Willems, R.J., Urwin, R. and Maiden, M.C. (2001) Multilocus sequence typing system for Campylobacter jejuni. JClin Microbiol 39, 14-23.

Dingle, K.E., McCarthy, N.D., Cody, A.J., Peto, T.E. and Maiden, M.C. (2008) Extended sequence typing of Campylobacter spp., United Kingdom. Emerg Infect Dis 14, 1620-1622.

56

Dingle, K.E., Van Den Braak, N., Colles, F.M., Price, L.J., Woodward, D.L., Rodgers, F.G., Endtz, H.P., Van Belkum, A. and Maiden, M.C. (2001) Sequence typing confirms that Campylobacter jejuni strains associated with Guillain-Barre and Miller-Fisher syndromes are of diverse genetic lineage, serotype, and flagella type. J Clin Microbiol39, 3346-3349.

Doorduyn, Y., VAN DEN Brandhof, W.E., VAN Duynhoven, Y.T., Breukink, B.J., Wagenaar, J.A. and VAN Pelt, W. (2010) Risk factors for indigenous Campylobacter jejuni and Campylobacter coli infections in The Netherlands: a case-control study. Epidemiol Infect, 1-14.

Doorduyn, Y., Van Pelt, W., Siezen, C.L., Van Der Horst, F., Van Duynhoven, Y.T., Hoebee, B. and Janssen, R. (2008) Novel insight in the association between salmonellosis or campylobacteriosis and chronic illness, and the role of host genetics in susceptibility to these diseases. Epidemiol Infect 136, 1225-1234.

Dorrell, N., Champion, O.L. and Wren, B.W. (2002) Microarray analysis of Campylobacter jejuni: to the guts of the problem! CompFunct Genomics 3, 338-341.

Doyle, M.P. and Roman, D.J. (1982) Prevalence and survival of Campylobacter jejuni in unpasteurized milk. Appl Environ Microbiol44, 1154-1158.

Duim, B., Ang, C.W., van Belkum, A., Rigter, A., van Leeuwen, N.W., Endtz, H.P. and Wagenaar, J.A. (2000) Amplified fragment length polymorphism analysis of Campylobacter jejuni strains isolated from chickens and from patients with gastroenteritis or Guillain-Barre or Miller Fisher syndrome. Appl Environ Microbiol 66, 3917-3923.

Duim, B., Wassenaar, T.M., Rigter, A. and Wagenaar, J. (1999) High-resolution genotyping of Campylobacter strains isolated from poultry and humans with amplified fragment length polymorphism fingerprinting. Appl Environ Microbiol 65, 2369-2375.

EFSA (2010a) Antimicrobial resistance in zoonotic and indicator bacteria from animals and foodin the European Union in 2004-2007. The EFSA Journal 8, 1309.

EFSA (2010b) The Community Summary Report on trends and sources of zoonosis and zoonotic agents and food-borne outbreaks in the European Union in 2008. The EFSA Journal, 104-111.

EFSA (2010c) Report of Task Force on Zoonoses Data Collection on the analysis of the baseline survey on the prevalence of Campylobacter in broiler batches and of Salmonella on broiler carcasses in the EU, 2008. The EFSA Journal 8(03):1503. The EFSA Journal 8, 1503.

57

EFSA (2007) Report of the Task Force of Zoonoses Data Collection including a proposal for a harmonizedmonitoring scheme of antimicrobial resistance in Salmonella in fowl (Gallus gallus), turkeys and pigs and Campylobacter jejuni and C. coli in broilers. The EFSA Journal, 1-46.

EFSA Panel on Biological Hazards (BIOHAZ) (2010) Scientific Opinion on Quantification of the risk posed by broiler meat to human campylobacteriosis in the EU. EFSA Journal 8, 1437-1526.

Ellis-Iversen, J., Cook, A.J., Smith, R.P., Pritchard, G.C. and Nielen, M. (2009a) Temporal patterns and risk factors for Escherichia coli O157 and Campylobacter spp, in young cattle. J Food Prot 72, 490-496.

Ellis-Iversen, J., Pritchard, G.C., Wooldridge, M. and Nielen, M. (2009b) Risk factors for Campylobacter jejuni and Campylobacter coli in young cattle on English and Welsh farms. Prev Vet Med 88, 42-48.

Endtz, H.P., Ruijs, G.J., van Klingeren, B., Jansen, W.H., van der Reyden, T. and Mouton, R.P. (1991) Quinolone resistance in campylobacter isolated from man and poultry following the introduction of fluoroquinolones in veterinary medicine. J Antimicrob Chemother 27, 199-208.

Engberg, J., Gerner-Smidt, P., Scheutz, F., Moller Nielsen, E., On, S.L. and Molbak, K. (1998) Water-borne Campylobacter jejuni infection in a Danish town---a 6-week continuous source outbreak. Clin Microbiol Infect 4, 648-656.

Engberg, J., Neimann, J., Nielsen, E.M., Aerestrup, F.M. and Fussing, V. (2004) Quinolone-resistant Campylobacter infections: risk factors and clinical consequences. Emerg Infect Dis 10, 1056-1063.

Englen, M.D., Hill, A.E., Dargatz, D.A., Ladely, S.R. and Fedorka-Cray, P.J. (2007) Prevalence and antimicrobial resistance of Campylobacter in US dairy cattle. J Appl Microbiol 102, 1570-1577.

Enokimoto, M., Kubo, M., Bozono, Y., Mieno, Y. and Misawa, N. (2007) Enumeration and identification of Campylobacter species in the liver and bile of slaughtered cattle. Int J Food Microbiol 118, 259-263.

Ethelberg, S., Simonsen, J., Gerner-Smidt, P., Olsen, K.E. and Molbak, K. (2005) Spatial distribution and registry-based case-control analysis of Campylobacter infections in Denmark, 1991-2001. Am J Epidemiol 162, 1008-1015.

EUCAST (2003) Determination of minimum inhibitory concentrations (MICs) of antibacterial agents by broth microdilution.EUCAST Discussion Document E. Def 2003, 5.1. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases, 1-10.

58

Evans, M.R., Roberts, R.J., Ribeiro, C.D., Gardner, D. and Kembrey, D. (1996) A milk-borne campylobacter outbreak following an educational farm visit. Epidemiol Infect 117, 457-462.

Fahey, T., Morgan, D., Gunneburg, C., Adak, G.K., Majid, F. and Kaczmarski, E. (1995) An outbreak of Campylobacter jejuni enteritis associated with failed milk pasteurisation. J Infect 31, 137-143.

Feodoroff, F.B., Lauhio, A.R., Sarna, S.J., Hänninen, M.L. and Rautelin, H.I. (2009) Severe diarrhoea caused by highly ciprofloxacin-susceptible Campylobacter isolates. Clin Microbiol Infect 15, 188-192.

Fitzgerald, C., Stanley, K., Andrew, S. and Jones, K. (2001) Use of pulsed-field gel electrophoresis and flagellin gene typing in identifying clonal groups of Campylobacter jejuni and Campylobacter coli in farm and clinical environments. Appl Environ Microbiol 67, 1429-1436.

Forbes, K.J., Gormley, F.J., Dallas, J.F., Labovitiadi, O., MacRae, M., Owen, R.J., Richardson, J., Strachan, N.J., Cowden, J.M., Ogden, I.D. and McGuigan, C.C. (2009) Campylobacter immunity and coinfection following a large outbreak in a farming community. J Clin Microbiol47, 111-116.

French, N., Barrigas, M., Brown, P., Ribiero, P., Williams, N., Leatherbarrow, H., Birtles, R., Bolton, E., Fearnhead, P. and Fox, A. (2005) Spatial epidemiology and natural population structure of Campylobacter jejuni colonizing a farmland ecosystem. EnvironMicrobiol 7, 1116-1126.

Gallay, A., De Valk, H., Cournot, M., Ladeuil, B., Hemery, C., Castor, C., Bon, F., Megraud, F., Le Cann, P., Desenclos, J.C. and Outbreak Investigation Team (2006) A large multi-pathogen waterborne community outbreak linked to faecal contamination of a groundwater system, France, 2000. Clin Microbiol Infect 12, 561-570.

Garcia, M.M., Lior, H., Stewart, R.B., Ruckerbauer, G.M., Trudel, J.R. and Skljarevski, A. (1985) Isolation, characterization, and serotyping of Campylobacter jejuni and Campylobacter coli from slaughter cattle. Appl Environ Microbiol 49, 667-672.

Garenaux, A., Guillou, S., Ermel, G., Wren, B., Federighi, M. and Ritz, M. (2008) Role of the Cj1371 periplasmic protein and the Cj0355c two-component regulator in the Campylobacter jejuni NCTC 11168 response to oxidative stress caused by paraquat. Res Microbiol159, 718-726.

Garrett, N., Devane, M.L., Hudson, J.A., Nicol, C., Ball, A., Klena, J.D., Scholes, P., Baker, M.G., Gilpin, B.J. and Savill, M.G. (2007) Statistical comparison of Campylobacter jejuni subtypes from human cases and environmental sources. J Appl Microbiol 103, 2113-2121.

59

Ghafir, Y., China, B., Dierick, K., De Zutter, L. and Daube, G. (2007) A seven-year survey of Campylobacter contamination in meat at different production stages in Belgium. Int J Food Microbiol 116, 111-120.

Gharst, G., Hanson, D. and Kathariou, S. (2006) Effect of direct culture versus selective enrichment on the isolation of thermophilic Campylobacter from feces of mature cattle at harvest. J Food Prot 69, 1024-1027.

Giacoboni, G.I., Itoh, K., Hirayama, K., Takahashi, E. and Mitsuoka, T. (1993) Comparison of fecal Campylobacter in calves and cattle of different ages and areas in Japan. J Vet Med Sci 55, 555-559.

Gillespie, I.A., O'Brien, S.J. and Bolton, F.J. (2009) Age patterns of persons with campylobacteriosis, England and Wales, 1990-2007. Emerg Infect Dis 15, 2046-2048.

Gilpin, B., Cornelius, A., Robson, B., Boxall, N., Ferguson, A., Nicol, C. and Henderson, T. (2006) Application of pulsed-field gel electrophoresis to identify potential outbreaks of campylobacteriosis in New Zealand. J Clin Microbiol 44, 406-412.

Gilpin, B.J., Scholes, P., Robson, B. and Savill, M.G. (2008a) The transmission of thermotolerant Campylobacter spp. to people living or working on dairy farms in New Zealand. Zoonoses Public Health 55, 352-360.

Gilpin, B.J., Thorrold, B., Scholes, P., Longhurst, R.D., Devane, M., Nicol, C., Walker, S., Robson, B. and Savill, M. (2008b) Comparison of Campylobacter jejuni genotypes from dairy cattle and human sources from the Matamata-Piako District of New Zealand. J Appl Microbiol 105, 1354-1360.

Grau, F. (1988) Campylobacter jejuni and Campylobacterhyointestinalis in the intestinal tract and on the carcasses of calves and cattle. J.Food Prot. 51, 857-861.

Grove-White, D.H., Leatherbarrow, A.J., Cripps, P.J., Diggle, P.J. and French, N.P. (2010) Temporal and farm-management-associated variation in the faecal-pat prevalence of Campylobacter jejuni in ruminants. Epidemiol Infect 138, 549-558.

Gupta, A., Nelson, J.M., Barrett, T.J., Tauxe, R.V., Rossiter, S.P., Friedman, C.R., Joyce, K.W., Smith, K.E., Jones, T.F., Hawkins, M.A., Shiferaw, B., Beebe, J.L., Vugia, D.J., Rabatsky-Ehr, T., Benson, J.A., Root, T.P., Angulo, F.J. and NARMS Working Group (2004) Antimicrobial resistance among Campylobacter strains, United States, 1997-2001. Emerg Infect Dis 10, 1102-1109.

Hakkinen, M. and Kaukonen, E. (2009) Campylobacters in Finnish broiler flocks: contamination level, diversity, and role of breeders in

60

transmission. 15th International workshop on Campylobacter, Helicobacter and related organisms., 79.

Hald, B. and Madsen, M. (1997) Healthy puppies and kittens as carriers of Campylobacter spp., with special reference to Campylobacter upsaliensis. J Clin Microbiol 35, 3351-3352.

Hald, B., Pedersen, K., Waino, M., Jorgensen, J.C. and Madsen, M. (2004) Longitudinal study of the excretion patterns of thermophilic Campylobacter spp. in young pet dogs in Denmark. J Clin Microbiol42, 2003-2012.

Hänninen, M.L., Haajanen, H., Pummi, T., Wermundsen, K., Katila, M.L., Sarkkinen, H., Miettinen, I. and Rautelin, H. (2003) Detection and typing of Campylobacter jejuni and Campylobacter coli and analysis of indicator organisms in three waterborne outbreaks in Finland. Appl Environ Microbiol 69, 1391-1396.

Hänninen, M.L., Niskanen, M. and Korhonen, L. (1998) Water as a reservoir for Campylobacter jejuni infection in cows studied by serotyping and pulsed-field gel electrophoresis (PFGE). ZentralblVeterinarmed B 45, 37-42.

Hänninen, M.L., Perko-Mäkelä, P., Pitkälä, A. and Rautelin, H. (2000) A three-year study of Campylobacter jejuni genotypes in humans with domestically acquired infections and in chicken samples from the Helsinki area. J Clin Microbiol 38, 1998-2000.

Hänninen, M.L., Perko-Mäkelä, P., Rautelin, H., Duim, B. and Wagenaar, J.A. (2001) Genomic relatedness within five common Finnish Campylobacter jejuni pulsed-field gel electrophoresis genotypes studied by amplified fragment length polymorphism analysis, ribotyping, and serotyping. Appl Environ Microbiol 67, 1581-1586.

Harrington, C.S., Thomson-Carter, F.M. and Carter, P.E. (1997) Evidence for recombination in the flagellin locus of Campylobacterjejuni: implications for the flagellin gene typing scheme. J Clin Microbiol 35, 2386-2392.

Helms, M., Simonsen, J., Olsen, K.E. and Molbak, K. (2005) Adverse health events associated with antimicrobial drug resistance in Campylobacter species: a registry-based cohort study. J Infect Dis191, 1050-1055.

Heuvelink, A.E., van Heerwaarden, C., Zwartkruis-Nahuis, A., Tilburg, J.J., Bos, M.H., Heilmann, F.G., Hofhuis, A., Hoekstra, T. and de Boer, E. (2009) Two outbreaks of campylobacteriosis associated with the consumption of raw cows' milk. Int J Food Microbiol 134, 70-74.

61

Hiett, K.L., Stintzi, A., Andacht, T.M., Kuntz, R.L. and Seal, B.S. (2008) Genomic differences between Campylobacter jejuni isolates identify surface membrane and flagellar function gene products potentially important for colonizing the chicken intestine. Funct Integr Genomics 8, 407-420.

Hoar, B.R., Atwill, E.R., Elmi, C., Utterback, W.W. and Edmondson, A.J. (1999) Comparison of fecal samples collected per rectum and off the ground for estimation of environmental contamination attributable to beef cattle. Am J Vet Res 60, 1352-1356.

Hofreuter, D., Novik, V. and Galan, J.E. (2008) Metabolic diversity in Campylobacter jejuni enhances specific tissue colonization. Cell Host Microbe 4, 425-433.

Hofreuter, D., Tsai, J., Watson, R.O., Novik, V., Altman, B., Benitez, M., Clark, C., Perbost, C., Jarvie, T., Du, L. and Galan, J.E. (2006) Unique features of a highly pathogenic Campylobacter jejuni strain. Infect Immun 74, 4694-4707.

Hong, J., Kim, J.M., Jung, W.K., Kim, S.H., Bae, W., Koo, H.C., Gil, J., Kim, M., Ser, J. and Park, Y.H. (2007) Prevalence and antibiotic resistance of Campylobacter spp. isolated from chicken meat, pork, and beef in Korea, from 2001 to 2006. J Food Prot 70, 860-866.

Hudson, J.A., Nicol, C., Wright, J., Whyte, R. and Hasell, S.K. (1999) Seasonal variation of Campylobacter types from human cases, veterinary cases, raw chicken, milk and water. J Appl Microbiol 87, 115-124.

Humphrey, T., O'Brien, S. and Madsen, M. (2007) Campylobacters as zoonotic pathogens: a food production perspective. Int J Food Microbiol 117, 237-257.

Humphrey, T.J. and Beckett, P. (1987) Campylobacter jejuni in dairy cows and raw milk. Epidemiol Infect 98, 263-269.

Hussain, I., Shahid Mahmood, M., Akhtar, M. and Khan, A. (2007) Prevalence of Campylobacter species in meat, milk and other food commodities in Pakistan. Food Microbiol 24, 219-222.

Inglis, G.D. and Kalischuk, L.D. (2004) Direct quantification of Campylobacter jejuni and Campylobacter lanienae in feces of cattle by real-time quantitative PCR. Appl Environ Microbiol 70, 2296-2306.

Inglis, G.D., Kalischuk, L.D. and Busz, H.W. (2004) Chronic shedding of Campylobacter species in beef cattle. J Appl Microbiol 97, 410-420.

Inglis, G.D., Kalischuk, L.D. and Busz, H.W. (2003) A survey of Campylobacter species shed in faeces of beef cattle using polymerase chain reaction. Can J Microbiol 49, 655-661.

62

Inglis, G.D., Kalischuk, L.D., Busz, H.W. and Kastelic, J.P. (2005) Colonization of cattle intestines by Campylobacter jejuni and Campylobacter lanienae. Appl Environ Microbiol 71, 5145-5153.

Ishihara, K., Yamamoto, T., Satake, S., Takayama, S., Kubota, S., Negishi, H., Kojima, A., Asai, T., Sawada, T., Takahashi, T. and Tamura, Y. (2006) Comparison of Campylobacter isolated from humans and food-producing animals in Japan. J Appl Microbiol 100, 153-160.

ISO (2006) International Standard ISO 10272-1. Microbiology of food and animal feeding stuffs – Horizontal method for detection and enumeration of Campylobacter spp. Part 1: Detection method.

Jacobs, B.C., Van Belkum, A. and Endtz, H.P. (2008) Guillain-Barré syndrome and Campylobacter infection. In ed. Nachamkin, I., Szymanski, C.M. and Blaser, M.J. pp. 245-261. Washington, DC, USA: ASM Press.

Jayarao, B.M., Donaldson, S.C., Straley, B.A., Sawant, A.A., Hegde, N.V. and Brown, J.L. (2006) A survey of foodborne pathogens in bulk tank milk and raw milk consumption among farm families in Pennsylvania. J Dairy Sci 89, 2451-2458.

Jayarao, B.M. and Henning, D.R. (2001) Prevalence of foodborne pathogens in bulk tank milk. J Dairy Sci 84, 2157-2162.

Johnsen, G., Zimmerman, K., Lindstedt, B.A., Vardund, T., Herikstad, H. and Kapperud, G. (2006) Intestinal carriage of Campylobacterjejuni and Campylobacter coli among cattle from south-western Norway and comparative genotyping of bovine and human isolates by amplified-fragment length polymorphism. Acta Vet Scand 48, 4.

Jones, P.H., Willis, A.T., Robinson, D.A., Skirrow, M.B. and Josephs, D.S. (1981) Campylobacter enteritis associated with the consumption of free school milk. J Hyg (Lond) 87, 155-162.

Jore, S., Viljugrein, H., Brun, E., Heier, B.T., Borck, B., Ethelberg, S., Hakkinen, M., Kuusi, M., Reiersen, J., Hansson, I., Engvall, E.O., Lofdahl, M., Wagenaar, J.A., van Pelt, W. and Hofshagen, M. (2010) Trends in Campylobacter incidence in broilers and humans in six European countries, 1997-2007. Prev Vet Med 93, 33-41.

Kahlmeter, G., Brown, D.F., Goldstein, F.W., MacGowan, A.P., Mouton, J.W., Osterlund, A., Rodloff, A., Steinbakk, M., Urbaskova, P. and Vatopoulos, A. (2003) European harmonization of MIC breakpoints for antimicrobial susceptibility testing of bacteria. JAntimicrob Chemother 52, 145-148.

Kalman, M., Szollosi, E., Czermann, B., Zimanyi, M., Szekeres, S. and Kalman, M. (2000) Milkborne campylobacter infection in Hungary. JFood Prot 63, 1426-1429.

63

Kapperud, G., Espeland, G., Wahl, E., Walde, A., Herikstad, H., Gustavsen, S., Tveit, I., Natas, O., Bevanger, L. and Digranes, A. (2003) Factors associated with increased and decreased risk of Campylobacter infection: a prospective case-control study in Norway. Am J Epidemiol 158, 234-242.

Kärenlampi, R., Rautelin, H., Hakkinen, M. and Hänninen, M.L. (2003) Temporal and geographical distribution and overlap of Penner heat-stable serotypes and pulsed-field gel electrophoresis genotypes of Campylobacter jejuni isolates collected from humans and chickens in Finland during a seasonal peak. J Clin Microbiol 41, 4870-4872.

Kärenlampi, R., Rautelin, H., Schönberg-Norio, D., Paulin, L. and Hänninen, M.L. (2007) Longitudinal study of Finnish Campylobacterjejuni and C. coli isolates from humans, using multilocus sequence typing, including comparison with epidemiological data and isolates from poultry and cattle. Appl Environ Microbiol 73, 148-155.

Karlyshev, A.V., Linton, D., Gregson, N.A., Lastovica, A.J. and Wren, B.W. (2000) Genetic and biochemical evidence of a Campylobacterjejuni capsular polysaccharide that accounts for Penner serotype specificity. Mol Microbiol 35, 529-541.

King, E.O. (1957) Human infections with Vibrio fetus and a closely related vibrio. J Infect Dis 101, 119-128.

Kovats, R.S., Edwards, S.J., Charron, D., Cowden, J., D'Souza, R.M., Ebi, K.L., Gauci, C., Gerner-Smidt, P., Hajat, S., Hales, S., Hernandez Pezzi, G., Kriz, B., Kutsar, K., McKeown, P., Mellou, K., Menne, B., O'Brien, S., van Pelt, W. and Schmid, H. (2005) Climate variability and campylobacter infection: an international study. Int J Biometeorol49, 207-214.

Kramer, J.M., Frost, J.A., Bolton, F.J. and Wareing, D.R. (2000) Campylobacter contamination of raw meat and poultry at retail sale: identification of multiple types and comparison with isolates from human infection. J Food Prot 63, 1654-1659.

Kwan, P.S., Barrigas, M., Bolton, F.J., French, N.P., Gowland, P., Kemp, R., Leatherbarrow, H., Upton, M. and Fox, A.J. (2008a) Molecular epidemiology of Campylobacter jejuni populations in dairy cattle, wildlife, and the environment in a farmland area. Appl Environ Microbiol 74, 5130-5138.

Kwan, P.S., Birtles, A., Bolton, F.J., French, N.P., Robinson, S.E., Newbold, L.S., Upton, M. and Fox, A.J. (2008b) Longitudinal study of the molecular epidemiology of Campylobacter jejuni in cattle on dairy farms. Appl Environ Microbiol 74, 3626-3633.

Lake, I.R., Gillespie, I.A., Bentham, G., Nichols, G.L., Lane, C., Adak, G.K. and Threlfall, E.J. (2009) A re-evaluation of the impact of

64

temperature and climate change on foodborne illness. Epidemiol Infect137, 1538-1547.

Lastovica, A.J. and Allos, B.M. (2008) Clinical significance of Campylobacter and related species other than Campylobacter jejuni and Campylobacter coli. In Campylobacter ed. Nachamkin, I., Szymanski, C.M. and Blaser, M.J. pp. 123-149. Washington, DC, USA: ASM Press.

Lee, M.D. and Newell, D.G. (2006) Campylobacter in poultry: filling an ecological niche. Avian Dis 50, 1-9.

Lehner, A., Schneck, C., Feierl, G., Pless, P., Deutz, A., Brandl, E. and Wagner, M. (2000) Epidemiologic application of pulsed-field gel electrophoresis to an outbreak of Campylobacter jejuni in an Austrian youth centre. Epidemiol Infect 125, 13-16.

Levesque, S., Frost, E., Arbeit, R.D. and Michaud, S. (2008) Multilocus sequence typing of Campylobacter jejuni isolates from humans, chickens, raw milk, and environmental water in Quebec, Canada. J Clin Microbiol 46, 3404-3411.

Levy, A.J. (1946) A gastroenteritis outbreak probably due to a bovine strain of vibrio. Yale J. Biol. Med. 18, 243.

Lindmark, H., Boqvist, S., Ljungstrom, M., Agren, P., Bjorkholm, B. and Engstrand, L. (2009) Risk factors for campylobacteriosis: an epidemiological surveillance study of patients and retail poultry. J Clin Microbiol 47, 2616-2619.

Lindmark, H., Harbom, B., Thebo, L., Andersson, L., Hedin, G., Osterman, B., Lindberg, T., Andersson, Y., Westoo, A. and Olsson Engvall, E. (2004) Genetic characterization and antibiotic resistance of Campylobacter jejuni isolated from meats, water, and humans in Sweden. J Clin Microbiol 42, 700-706.

Lior, H., Woodward, D.L., Edgar, J.A., Laroche, L.J. and Gill, P. (1982) Serotyping of Campylobacter jejuni by slide agglutination based on heat-labile antigenic factors. J Clin Microbiol 15, 761-768.

Little, C.L., Richardson, J.F., Owen, R.J., de Pinna, E. and Threlfall, E.J. (2008) Prevalence, characterisation and antimicrobial resistance of Campylobacter and Salmonella in raw poultrymeat in the UK, 2003-2005. Int J Environ Health Res 18, 403-414.

Louis, V.R., Gillespie, I.A., O'Brien, S.J., Russek-Cohen, E., Pearson, A.D. and Colwell, R.R. (2005) Temperature-driven Campylobacterseasonality in England and Wales. Appl Environ Microbiol 71, 85-92.

Lovett, J., Francis, D.W. and Hunt, J.M. (1983) Isolation of Campylobacter jejuni from raw milk. Appl Environ Microbiol 46, 459-462.

65

Lukinmaa, S., Nakari, U.M., Eklund, M. and Siitonen, A. (2004) Application of molecular genetic methods in diagnostics and epidemiology of food-borne bacterial pathogens. APMIS 112, 908-929.

Madden, R.H., Murray, K.A. and Gilmour, A. (2007) Carriage of four bacterial pathogens by beef cattle in Northern Ireland at time of slaughter. Lett Appl Microbiol 44, 115-119.

Manning, G., Dowson, C.G., Bagnall, M.C., Ahmed, I.H., West, M. and Newell, D.G. (2003) Multilocus sequence typing for comparison of veterinary and human isolates of Campylobacter jejuni. Appl Environ Microbiol 69, 6370-6379.

Matsumoto, N., Taniwaki, T., Kinuta, M. and Murase, T. (2008) Isolation of Campylobacter jejuni and coliform bacilli from bile and liver obtained from slaughter cattle in Western Japan. J Food Prot 71, 1228-1231.

McCarthy, N.D., Colles, F.M., Dingle, K.E., Bagnall, M.C., Manning, G., Maiden, M.C. and Falush, D. (2007) Host-associated genetic import in Campylobacter jejuni. Emerg Infect Dis 13, 267-272.

McNaughton, R.D., Leyland, R. and Mueller, L. (1982) Outbreak of Campylobacter enteritis due to consumption of raw milk. Can Med Assoc J 126, 657-658.

Medeiros, D.T., Sattar, S.A., Farber, J.M. and Carrillo, C.D. (2008) Occurrence of Campylobacter spp. in raw and ready-to-eat foods and in a Canadian food service operation. J Food Prot 71, 2087-2093.

Meinersmann, R.J., Helsel, L.O., Fields, P.I. and Hiett, K.L. (1997) Discrimination of Campylobacter jejuni isolates by fla gene sequencing. J Clin Microbiol 35, 2810-2814.

Meinersmann, R.J., Phillips, R.W., Hiett, K.L. and Fedorka-Cray, P. (2005) Differentiation of Campylobacter populations as demonstrated by flagellin short variable region sequences. Appl Environ Microbiol71, 6368-6374.

Michaud, S., Menard, S. and Arbeit, R.D. (2004) Campylobacteriosis, Eastern Townships, Quebec. Emerg Infect Dis 10, 1844-1847.

Michaud, S., Menard, S., Gaudreau, C. and Arbeit, R.D. (2001) Comparison of SmaI-defined genotypes of Campylobacter jejuni examined by KpnI: a population-based study. J Med Microbiol 50, 1075-1081.

Milnes, A.S., Stewart, I., Clifton-Hadley, F.A., Davies, R.H., Newell, D.G., Sayers, A.R., Cheasty, T., Cassar, C., Ridley, A., Cook, A.J., Evans, S.J., Teale, C.J., Smith, R.P., McNally, A., Toszeghy, M., Futter, R., Kay, A. and Paiba, G.A. (2008) Intestinal carriage of verocytotoxigenic Escherichia coli O157, Salmonella, thermophilic

66

Campylobacter and Yersinia enterocolitica, in cattle, sheep and pigs at slaughter in Great Britain during 2003. Epidemiol Infect 136, 739-751.

Minihan, D., Whyte, P., O'Mahony, M., Fanning, S., McGill, K. and Collins, J.D. (2004) Campylobacter spp. in Irish feedlot cattle: a longitudinal study involving pre-harvest and harvest phases of the food chain. J Vet Med B Infect Dis Vet Public Health 51, 28-33.

Morgan, D., Gunneberg, C., Gunnell, D., Healing, T.D., Lamerton, S., Soltanpoor, N., Lewis, D.A. and White, D.G. (1994) An outbreak of Campylobacter infection associated with the consumption of unpasteurised milk at a large festival in England. Eur J Epidemiol 10, 581-585.

Mullner, P., Collins-Emerson, J.M., Midwinter, A.C., Carter, P., Spencer, S.E., van der Logt, P., Hathaway, S. and French, N.P. (2010a) Molecular epidemiology of Campylobacter jejuni in a geographically isolated country with a uniquely structured poultry industry. Appl Environ Microbiol 76, 2145-2154.

Mullner, P., Shadbolt, T., Collins-Emerson, J.M., Midwinter, A.C., Spencer, S.E., Marshall, J., Carter, P.E., Campbell, D.M., Wilson, D.J., Hathaway, S., Pirie, R. and French, N.P. (2010b) Molecular and spatial epidemiology of human campylobacteriosis: source association and genotype-related risk factors. Epidemiol Infect, 1-12.

Mullner, P., Spencer, S.E., Wilson, D.J., Jones, G., Noble, A.D., Midwinter, A.C., Collins-Emerson, J.M., Carter, P., Hathaway, S. and French, N.P. (2009) Assigning the source of human campylobacteriosis in New Zealand: A comparative genetic and epidemiological approach. Infect Genet Evol.

Nakari, U.M., Huovinen, E., Kuusi, M. and Siitonen, A. (2010) Population-based surveillance study of Campylobacter infections in Finland. Epidemiol Infect, 1-7.

Nakari, U.M., Laaksonen, K., Korkeila, M. and Siitonen, A. (2005) Comparative typing of Campylobacter jejuni by heat-stable serotyping and PCR-based restriction fragment length polymorphism analysis. JClin Microbiol 43, 1166-1170.

National Institute for Health and Welfare (2009) Infectious diseases in Finland in 2008.

National Public Health Institute (2005) Infectious diseases in Finland 1995-2004.

NCFA (Nordic Committee on Food Analysis) (2007) Thermotolerant Campylobacter. Detection, semi-quantitative and quantitative determination in foods and drinking water, Method No. 119, 3.

67

Neimann, J., Engberg, J., Molbak, K. and Wegener, H.C. (2003) A case-control study of risk factors for sporadic campylobacter infections in Denmark. Epidemiol Infect 130, 353-366.

Nicholson, F.A., Groves, S.J. and Chambers, B.J. (2005) Pathogen survival during livestock manure storage and following land application. Bioresour Technol 96, 135-143.

Nielsen, E.M. (2002) Occurrence and strain diversity of thermophilic campylobacters in cattle of different age groups in dairy herds. LettAppl Microbiol 35, 85-89.

Nielsen, E.M., Engberg, J. and Madsen, M. (1997) Distribution of serotypes of Campylobacter jejuni and C. coli from Danish patients, poultry, cattle and swine. FEMS Immunol Med Microbiol 19, 47-56.

Nygård, K., Andersson, Y., Rottingen, J.A., Svensson, A., Lindback, J., Kistemann, T. and Giesecke, J. (2004) Association between environmental risk factors and campylobacter infections in Sweden. Epidemiol Infect 132, 317-325.

Nylen, G., Dunstan, F., Palmer, S.R., Andersson, Y., Bager, F., Cowden, J., Feierl, G., Galloway, Y., Kapperud, G., Megraud, F., Molbak, K., Petersen, L.R. and Ruutu, P. (2002) The seasonal distribution of campylobacter infection in nine European countries and New Zealand. Epidemiol Infect 128, 383-390.

Olson, C.K., Ethelberg, S., van Pelt, W. and Tauxe, R.V. (2008) Epidemiology of Campylobacter jejuni infections in industrialized nations. In ed. Nachamkin, I., Szymanski, C.M. and Blaser, M. pp. 163-189. Washington DC, USA: ASM Press.

Oosterom, J., de Wilde, G.J.A., de Boer, E., de Blaauw, L.H. and Karman, H. (1983) Survival of Campylobacter during poultry processing and pig slaughtering. Journal of food protection 46, 702-706.

Oporto, B., Esteban, J.I., Aduriz, G., Juste, R.A. and Hurtado, A. (2007) Prevalence and strain diversity of thermophilic campylobacters in cattle, sheep and swine farms. J Appl Microbiol 103, 977-984.

Parisi, A., Lanzilotta, S.G., Addante, N., Normanno, G., Di Modugno, G., Dambrosio, A. and Montagna, C.O. (2007) Prevalence, molecular characterization and antimicrobial resistance of thermophilic campylobacter isolates from cattle, hens, broilers and broiler meat in south-eastern Italy. Vet Res Commun 31, 113-123.

Parkhill, J., Wren, B.W., Mungall, K., Ketley, J.M., Churcher, C., Basham, D., Chillingworth, T., Davies, R.M., Feltwell, T., Holroyd, S., Jagels, K., Karlyshev, A.V., Moule, S., Pallen, M.J., Penn, C.W., Quail, M.A., Rajandream, M.A., Rutherford, K.M., van Vliet, A.H., Whitehead, S. and Barrell, B.G. (2000) The genome sequence of the

68

food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665-668.

Paton, A.W., Srimanote, P., Talbot, U.M., Wang, H. and Paton, J.C. (2004) A new family of potent AB(5) cytotoxins produced by Shiga toxigenic Escherichia coli. J Exp Med 200, 35-46.

Patrick, M.E., Christiansen, L.E., Waino, M., Ethelberg, S., Madsen, H. and Wegener, H.C. (2004) Effects of climate on incidence of Campylobacter spp. in humans and prevalence in broiler flocks in Denmark. Appl Environ Microbiol 70, 7474-7480.

Penner, J.L. and Hennessy, J.N. (1980) Passive hemagglutination technique for serotyping Campylobacter fetus subsp. jejuni on the basis of soluble heat-stable antigens. J Clin Microbiol 12, 732-737.

Penner, J.L., Hennessy, J.N. and Congi, R.V. (1983) Serotyping of Campylobacter jejuni and Campylobacter coli on the basis of thermostable antigens. Eur J Clin Microbiol 2, 378-383.

Perko-Mäkelä, P., Hakkinen, M., Honkanen-Buzalski, T. and Hänninen, M.L. (2002) Prevalence of campylobacters in chicken flocks during the summer of 1999 in Finland. Epidemiol Infect 129, 187-192.

Peterson, M.C. (2003) Campylobacter jejuni enteritis associated with consumption of raw milk. J Environ Health 65, 20-1, 24, 26.

Pezzotti, G., Serafin, A., Luzzi, I., Mioni, R., Milan, M. and Perin, R. (2003) Occurrence and resistance to antibiotics of Campylobacterjejuni and Campylobacter coli in animals and meat in northeastern Italy. Int J Food Microbiol 82, 281-287.

Phillips, D., Jordan, D., Morris, S., Jenson, I. and Sumner, J. (2008) A national survey of the microbiological quality of retail raw meats in Australia. J Food Prot 71, 1232-1236.

Pitkänen, T., Miettinen, I.T., Nakari, U.M., Takkinen, J., Nieminen, K., Siitonen, A., Kuusi, M., Holopainen, A. and Hänninen, M.L. (2008) Faecal contamination of a municipal drinking water distribution system in association with Campylobacter jejuni infections. J Water Health 6, 365-376.

Pradhan, A.K., Van Kessel, J.S., Karns, J.S., Wolfgang, D.R., Hovingh, E., Nelen, K.A., Smith, J.M., Whitlock, R.H., Fyock, T., Ladely, S., Fedorka-Cray, P.J. and Schukken, Y.H. (2009) Dynamics of endemic infectious diseases of animal and human importance on three dairy herds in the northeastern United States. J Dairy Sci 92, 1811-1825.

Ragimbeau, C., Schneider, F., Losch, S., Even, J. and Mossong, J. (2008) Multilocus sequence typing, pulsed-field gel electrophoresis,

69

and fla short variable region typing of clonal complexes of Campylobacter jejuni strains of human, bovine, and poultry origins in Luxembourg. Appl Environ Microbiol 74, 7715-7722.

Rahimi, E., Ameri, M. and Kazemeini, H.R. (2010) Prevalence and antimicrobial resistance of Campylobacter species isolated from raw camel, beef, lamb, and goat meat in Iran. Foodborne Pathog Dis 7, 443-447.

Rautelin, H. and Hänninen, M.L. (1999) Comparison of a commercial test for serotyping heat-stable antigens of Campylobacter jejuni with genotyping by pulsed-field gel electrophoresis. J Med Microbiol 48, 617-621.

Rautelin, H., Vierikko, A., Hänninen, M.L. and Vaara, M. (2003) Antimicrobial susceptibilities of Campylobacter strains isolated from Finnish subjects infected domestically or from those infected abroad. Antimicrob Agents Chemother 47, 102-105.

Reich, F., Atanassova, V., Haunhorst, E. and Klein, G. (2008) The effects of Campylobacter numbers in caeca on the contamination of broiler carcasses with Campylobacter. Int J Food Microbiol 127, 116-120.

Ribot, E.M., Fitzgerald, C., Kubota, K., Swaminathan, B. and Barrett, T.J. (2001) Rapid pulsed-field gel electrophoresis protocol for subtyping of Campylobacter jejuni. J Clin Microbiol 39, 1889-1894.

Robinson, D.A., Edgar, W.J., Gibson, G.L., Matchett, A.A. and Robertson, L. (1979) Campylobacter enteritis associated with consumption of unpasteurised milk. Br Med J 1, 1171-1173.

Robinson, D.A. and Jones, D.M. (1981) Milk-borne campylobacter infection. Br Med J (Clin Res Ed) 282, 1374-1376.

Rosenquist, H., Boysen, L., Galliano, C., Nordentoft, S., Ethelberg, S. and Borck, B. (2009) Danish strategies to control Campylobacter in broilers and broiler meat: facts and effects. Epidemiol Infect 137, 1742-1750.

Rotariu, O., Dallas, J.F., Ogden, I.D., MacRae, M., Sheppard, S.K., Maiden, M.C., Gormley, F.J., Forbes, K.J. and Strachan, N.J. (2009) Spatiotemporal homogeneity of Campylobacter subtypes from cattle and sheep across northeastern and southwestern Scotland. ApplEnviron Microbiol 75, 6275-6281.

Sails, A.D., Swaminathan, B. and Fields, P.I. (2003) Utility of multilocus sequence typing as an epidemiological tool for investigation of outbreaks of gastroenteritis caused by Campylobacterjejuni. J Clin Microbiol 41, 4733-4739.

70

Sammarco, M.L., Ripabelli, G., Fanelli, I., Grasso, G.M. and Tamburro, M. (2010) Prevalence and biomolecular characterization of Campylobacter spp. isolated from retail meat. J Food Prot 73, 720-728.

Sandberg, M., Nygård, K., Meldal, H., Valle, P.S., Kruse, H. and Skjerve, E. (2006) Incidence trend and risk factors for campylobacter infections in humans in Norway. BMC Public Health 6, 179.

Schildt, M., Savolainen, S. and Hänninen, M.L. (2006) Long-lasting Campylobacter jejuni contamination of milk associated with gastrointestinal illness in a farming family. Epidemiol Infect 134, 401-405.

Schönberg-Norio, D., Mattila, L., Lauhio, A., Katila, M.L., Kaukoranta, S.S., Koskela, M., Pajarre, S., Uksila, J., Eerola, E., Sarna, S. and Rautelin, H. (2009) Patient-reported complications associated with Campylobacter jejuni infection. Epidemiol Infect, 1-8.

Schönberg-Norio, D., Sarna, S., Hänninen, M.L., Katila, M.L., Kaukoranta, S.S. and Rautelin, H. (2006) Strain and host characteristics of Campylobacter jejuni infections in Finland. ClinMicrobiol Infect 12, 754-760.

Schönberg-Norio, D., Takkinen, J., Hänninen, M.L., Katila, M.L., Kaukoranta, S.S., Mattila, L. and Rautelin, H. (2004) Swimming and Campylobacter infections. Emerg Infect Dis 10, 1474-1477.

Schouls, L.M., Reulen, S., Duim, B., Wagenaar, J.A., Willems, R.J., Dingle, K.E., Colles, F.M. and Van Embden, J.D. (2003) Comparative genotyping of Campylobacter jejuni by amplified fragment length polymorphism, multilocus sequence typing, and short repeat sequencing: strain diversity, host range, and recombination. J Clin Microbiol 41, 15-26.

Schwarz, S., Silley, P., Simjee, S., Woodford, N., van Duijkeren, E., Johnson, A.P. and Gaastra, W. (2010) Editorial: assessing the antimicrobial susceptibility of bacteria obtained from animals. JAntimicrob Chemother 65, 601-604.

Sebald, E.R. and Véron, M. (1963) Teneur en bases da lÀDN et classification des vibrions. Annales de L'institut Pasteur (Paris) 105, 897-910.

Sheppard, S.K., Dallas, J.F., MacRae, M., McCarthy, N.D., Sproston, E.L., Gormley, F.J., Strachan, N.J., Ogden, I.D., Maiden, M.C. and Forbes, K.J. (2009) Campylobacter genotypes from food animals, environmental sources and clinical disease in Scotland 2005/6. Int J Food Microbiol 134, 96-103.

Skirrow, M.B. (2006) John McFadyean and the centenary of the first isolation of Campylobacter species. Clin Infect Dis 43, 1213-1217.

71

Skirrow, M.B. (1977) Campylobacter enteritis: a "new" disease. BrMed J 2, 9-11.

Smith, T. and Taylor, M.S. (1919) Some morphological and biochemical characteristics of the spirilla (Vibrio fetus n. sp.) associated with disease of the fetal membranes in cattle. J Exp Med, 299-312.

Sopwith, W., Ashton, M., Frost, J.A., Tocque, K., O'Brien, S., Regan, M. and Syed, Q. (2003) Enhanced surveillance of Campylobacter infection in the North West of England 1997-1999. J Infect 46, 35-45.

Stafford, R.J., Schluter, P., Kirk, M., Wilson, A., Unicomb, L., Ashbolt, R., Gregory, J. and OzFoodNet Working Group (2007) A multi-centre prospective case-control study of campylobacter infection in persons aged 5 years and older in Australia. Epidemiol Infect 135, 978-988.

Stafford, R.J., Schluter, P.J., Wilson, A.J., Kirk, M.D., Hall, G., Unicomb, L. and OzFoodNet Working Group (2008) Population-attributable risk estimates for risk factors associated with Campylobacter infection, Australia. Emerg Infect Dis 14, 895-901.

Stanley, K. and Jones, K. (2003) Cattle and sheep farms as reservoirs of Campylobacter. J Appl Microbiol 94 Suppl, 104S-113S.

Stanley, K.N., Wallace, J.S., Currie, J.E., Diggle, P.J. and Jones, K. (1998) The seasonal variation of thermophilic campylobacters in beef cattle, dairy cattle and calves. J Appl Microbiol 85, 472-480.

Stark, K., Niedrig, M., Biederbick, W., Merkert, H. and Hacker, J. (2009) Climate changes and emerging diseases. What new infectious diseases and health problem can be expected? BundesgesundheitsblattGesundheitsforschung Gesundheitsschutz 52, 699-714.

Stern, N.J., Hiett, K.L., Alfredsson, G.A., Kristinsson, K.G., Reiersen, J., Hardardottir, H., Briem, H., Gunnarsson, E., Georgsson, F., Lowman, R., Berndtson, E., Lammerding, A.M., Paoli, G.M. and Musgrove, M.T. (2003) Campylobacter spp. in Icelandic poultry operations and human disease. Epidemiol Infect 130, 23-32.

Strachan, N.J., Gormley, F.J., Rotariu, O., Ogden, I.D., Miller, G., Dunn, G.M., Sheppard, S.K., Dallas, J.F., Reid, T.M., Howie, H., Maiden, M.C. and Forbes, K.J. (2009) Attribution of Campylobacter infections in northeast Scotland to specific sources by use of multilocus sequence typing. J Infect Dis 199, 1205-1208.

Studahl, A. and Andersson, Y. (2000) Risk factors for indigenous campylobacter infection: a Swedish case-control study. Epidemiol Infect 125, 269-275.

72

Svedhem, A., Kaijser, B. and Sjogren, E. (1981) The occurrence of Campylobacter jejuni in fresh food and survival under different conditions. J Hyg (Lond) 87, 421-425.

Taboada, E.N., Acedillo, R.R., Carrillo, C.D., Findlay, W.A., Medeiros, D.T., Mykytczuk, O.L., Roberts, M.J., Valencia, C.A., Farber, J.M. and Nash, J.H. (2004) Large-scale comparative genomics meta-analysis of Campylobacter jejuni isolates reveals low level of genome plasticity. J Clin Microbiol 42, 4566-4576.

Tam, C.C., Higgins, C.D., Neal, K.R., Rodrigues, L.C., Millership, S.E., O'Brien, S.J. and Campylobacter Case-Control Study Group (2009) Chicken consumption and use of acid-suppressing medications as risk factors for Campylobacter enteritis, England. Emerg Infect Dis15, 1402-1408.

Tam, C.C., Rodrigues, L.C., O'Brien, S.J. and Hajat, S. (2006) Temperature dependence of reported Campylobacter infection in England, 1989-1999. Epidemiol Infect 134, 119-125.

Tenover, F.C., Arbeit, R.D., Goering, R.V., Mickelsen, P.A., Murray, B.E., Persing, D.H. and Swaminathan, B. (1995) Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 33, 2233-2239.

Teunis, P., Van den Brandhof, W., Nauta, M., Wagenaar, J., Van den Kerkhof, H. and Van Pelt, W. (2005) A reconsideration of the Campylobacter dose-response relation. Epidemiol Infect 133, 583-592.

Tribble, D.R., Baqar, S., Scott, D.A., Oplinger, M.L., Trespalacios, F., Rollins, D., Walker, R.I., Clements, J.D., Walz, S., Gibbs, P., Burg, E.F.,3rd, Moran, A.P., Applebee, L. and Bourgeois, A.L. (2010) Assessment of the duration of protection in Campylobacter jejuni experimental infection in humans. Infect Immun 78, 1750-1759.

Unicomb, L.E., Dalton, C.B., Gilbert, G.L., Becker, N.G. and Patel, M.S. (2008) Age-specific risk factors for sporadic Campylobacter infection in regional Australia. Foodborne Pathog Dis 5, 79-85.

Unicomb, L.E., Fullerton, K.E., Kirk, M.D. and Stafford, R.J. (2009) Outbreaks of campylobacteriosis in Australia, 2001 to 2006. Foodborne Pathog Dis 6, 1241-1250.

Unicomb, L.E., O'Reilly, L.C., Kirk, M.D., Stafford, R.J., Smith, H.V., Becker, N.G., Patel, M.S., Gilbert, G.L. and Australian Campylobacter Subtyping Study Group (2008) Risk factors for infection with Campylobacter jejuni flaA genotypes. Epidemiol Infect 136, 1480-1491.

Vally, H., Hall, G., Scallan, E., Kirk, M.D. and Angulo, F.J. (2009) Higher rate of culture-confirmed Campylobacter infections in

73

Australia than in the USA: is this due to differences in healthcare-seeking behaviour or stool culture frequency? Epidemiol Infect 137, 1751-1758.

van Hees, B.C., Veldman-Ariesen, M.J., de Jongh, B.M., Tersmette, M. and van Pelt, W. (2007) Regional and seasonal differences in incidence and antibiotic resistance of Campylobacter from a nationwide surveillance study in The Netherlands: an overview of 2000-2004. Clin Microbiol Infect 13, 305-310.

Vellinga, A. and Van Loock, F. (2002) The dioxin crisis as experiment to determine poultry-related Campylobacter enteritis. Emerg Infect Dis8, 19-22.

Vierikko, A., Hänninen, M.L., Siitonen, A., Ruutu, P. and Rautelin, H. (2004) Domestically acquired Campylobacter infections in Finland. Emerg Infect Dis 10, 127-130.

Vindigni, S.M., Srijan, A., Wongstitwilairoong, B., Marcus, R., Meek, J., Riley, P.L. and Mason, C. (2007) Prevalence of foodborne microorganisms in retail foods in Thailand. Foodborne Pathog Dis 4, 208-215.

Wang, H., Paton, J.C. and Paton, A.W. (2007) Pathologic changes in mice induced by subtilase cytotoxin, a potent new Escherichia coli AB5 toxin that targets the endoplasmic reticulum. J Infect Dis 196, 1093-1101.

Wareing, D.R., Ure, R., Colles, F.M., Bolton, F.J., Fox, A.J., Maiden, M.C. and Dingle, K.E. (2003) Reference isolates for the clonal complexes of Campylobacter jejuni. Lett Appl Microbiol 36, 106-110.

Wassenaar, T.M. and Newell, D.G. (2000) Genotyping of Campylobacter spp. Appl Environ Microbiol 66, 1-9.

Wesley, I.V., Wells, S.J., Harmon, K.M., Green, A., Schroeder-Tucker, L., Glover, M. and Siddique, I. (2000) Fecal shedding of Campylobacter and Arcobacter spp. in dairy cattle. Appl Environ Microbiol 66, 1994-2000.

White, A.N., Kinlin, L.M., Johnson, C., Spain, C.V., Ng, V. and Fisman, D.N. (2009) Environmental determinants of campylobacteriosis risk in Philadelphia from 1994 to 2007. Ecohealth6, 200-208.

WHO (2007) Report of the Second WHO Expert Meeting: Critically important antimicrobials for human medicine: Categorization for the development of risk management strategies to contain antimicrobial resistance due to non-human antimicrobial use. http://www.who.int/foodborne_disease/resistance/antimicrobials_humans.pdf.

74

Whyte, P., McGill, K., Cowley, D., Madden, R.H., Moran, L., Scates, P., Carroll, C., O'Leary, A., Fanning, S., Collins, J.D., McNamara, E., Moore, J.E. and Cormican, M. (2004) Occurrence of Campylobacter in retail foods in Ireland. Int J Food Microbiol 95, 111-118.

Wieland, B., Regula, G., Danuser, J., Wittwer, M., Burnens, A.P., Wassenaar, T.M. and Stark, K.D. (2005) Campylobacter spp. in dogs and cats in Switzerland: risk factor analysis and molecular characterization with AFLP. J Vet Med B Infect Dis Vet Public Health52, 183-189.

Wilson, D.J., Gabriel, E., Leatherbarrow, A.J., Cheesbrough, J., Gee, S., Bolton, E., Fox, A., Fearnhead, P., Hart, C.A. and Diggle, P.J. (2008) Tracing the source of campylobacteriosis. PLoS Genet 4, e1000203.

Wingstrand, A., Neimann, J., Engberg, J., Nielsen, E.M., Gerner-Smidt, P., Wegener, H.C. and Molbak, K. (2006) Fresh chicken as main risk factor for campylobacteriosis, Denmark. Emerg Infect Dis12, 280-285.

Wistrom, J. and Norrby, S.R. (1995) Fluoroquinolones and bacterial enteritis, when and for whom? J Antimicrob Chemother 36, 23-39.

Wong, T.L., Hollis, L., Cornelius, A., Nicol, C., Cook, R. and Hudson, J.A. (2007) Prevalence, numbers, and subtypes of Campylobacterjejuni and Campylobacter coli in uncooked retail meat samples. JFood Prot 70, 566-573.

Workman, S.N., Mathison, G.E. and Lavoie, M.C. (2005) Pet dogs and chicken meat as reservoirs of Campylobacter spp. in Barbados. J Clin Microbiol 43, 2642-2650.

Yang, C., Jiang, Y., Huang, K., Zhu, C. and Yin, Y. (2003) Application of real-time PCR for quantitative detection of Campylobacter jejuni in poultry, milk and environmental water. FEMSImmunol Med Microbiol 38, 265-271.

Zhao, C., Ge, B., De Villena, J., Sudler, R., Yeh, E., Zhao, S., White, D.G., Wagner, D. and Meng, J. (2001) Prevalence of Campylobacter spp., Escherichia coli, and Salmonella serovars in retail chicken, turkey, pork, and beef from the Greater Washington, D.C., area. ApplEnviron Microbiol 67, 5431-5436.

Zhao, T., Doyle, M.P. and Berg, D.E. (2000) Fate of Campylobacter jejuni in butter. J Food Prot 63, 120-122.

75

I

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, May 2007, p. 3232–3238 Vol. 73, No. 100099-2240/07/$08.00�0 doi:10.1128/AEM.02579-06Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Prevalence of Campylobacter spp. in Cattle in Finland and AntimicrobialSusceptibilities of Bovine Campylobacter jejuni Strains�

Marjaana Hakkinen,1* Helmi Heiska,1 and Marja-Liisa Hanninen2

Finnish Food Safety Authority, Mustialankatu 3, Helsinki FI-00790, Finland,1 and Department of Food andEnvironmental Hygiene, University of Helsinki, Helsinki FI-00014, Finland2

Received 9 November 2006/Accepted 12 March 2007

The study investigated the prevalence of Campylobacter spp. in Finnish cattle at slaughter and carcasscontamination after slaughter. During the period January to December 2003, bovine rectal fecal samples (n �952) and carcass surface samples (n � 948) from 12 out of 15 Finnish slaughterhouses were examined. In total,campylobacters were detected in 31.1% of fecal samples and in 3.5% of carcass surface samples. Campylobacterjejuni was isolated from 19.5%, Campylobacter coli from 2.2%, and presumptive Campylobacter hyointestinalisfrom 10.8% of fecal samples. Campylobacters were detected in 4.4% and 37.4% of the fecal samples examinedboth by direct culture and by enrichment (n � 730), respectively, suggesting a low level of campylobacters inthe intestinal content. A slightly increasing trend was observed in the overall prevalence of campylobacterstowards the end of summer and autumn. Seventeen different serotypes were detected among the fecal C. jejuniisolates using a set of 25 commercial antisera for serotyping heat-stable antigens (Penner) of C. jejuni bypassive hemagglutination. The predominant serotypes, Pen2 and Pen4-complex, were isolated from 52% of thefecal samples. Subtyping by pulsed-field gel electrophoresis (SmaI) yielded 56 and 20 subtypes out of 330 fecaland 70 carcass C. jejuni isolates, respectively. MICs of ampicillin, enrofloxacin, erythromycin, gentamicin,nalidixic acid, and oxytetracycline for 187 C. jejuni isolates were determined using a commercial brothmicrodilution method. Sixteen (9%) of the isolates were resistant to at least one of the antimicrobials tested.Resistance to nalidixic acid was most commonly detected (6%). No multiresistance was observed.

Over the last 20 years thermophilic campylobacters havebecome the most important human bacterial pathogens inmost western European countries (55a). In Finland the num-ber of reported cases has shown an increasing trend over thelast 10 years apart from a slight decrease from 2002 to 2003(35). In Finland, during the seasonal peak from June to Sep-tember in 2003 approximately 40% of the cases were of Finnishorigin (53).

Poultry is generally considered to be the most importantsingle reservoir for campylobacters, mainly Campylobacter je-juni. However, there is some evidence based on the temporaloccurrence of serotypes and genotypes shared by humans andpoultry and on weekly data for poultry and human isolates thatsuggests that there is a common source of campylobactersinstead of direct poultry-human transmission (28, 32). In ad-dition, genotyping data on campylobacters of human and ani-mal origin have raised the question of whether the role ofpoultry as a source of campylobacter infections has been over-estimated (21, 40, 48).

Cattle are also common carriers of campylobacters (23, 25,49). However, beef is not considered to be an important vehi-cle of transmission in human infections, because campy-lobacters are not commonly detected on carcasses or in beef.In surveys of retail beef only 0 to 5% of the samples have testedpositive for campylobacters (42, 50, 55). Instead, the impor-tance of raw milk as a risk factor for human campylobacteriosishas been recognized in epidemiological studies (33, 51), and

consumption of unpasteurized milk has been associated withcampylobacter infections in several outbreaks (12, 30, 47, 51).The environmental load of campylobacters in cattle manuremay be a more significant factor in the transmission of infec-tions than contaminated milk or beef (36, 39).

Antimicrobial treatment is not usually required for humancampylobacter infections. In cases with severe or prolongedsymptoms macrolides or fluoroquinolones have been recom-mended as treatment. Since the 1990s the increasing resistanceof campylobacters to antibiotics, especially to fluoroquinolo-nes, has been reported both among animal isolates and amongisolates from human infections (10, 18). Because person-to-person transmission of campylobacters is uncommon and in-fections are frequently acquired from foods of animal origin,the use of antimicrobials in production animals has been sug-gested as the cause of the increase in resistance (3, 41). InFinland, products containing macrolides and fluoroquinolonesare authorized for bovine use, but their use is limited.

The objective of the present study was to elucidate the roleof Finnish cattle as a potential reservoir for thermophiliccampylobacters and as a source of antibiotic-resistant C. jejuni.

MATERIALS AND METHODS

Sampling. Rectal fecal samples (n � 952) and carcass surface samples (n �948) from clinically healthy cattle were collected in 12 slaughterhouses in Finlandduring the period January to December 2003. Sampling was carried out weekly,every second week, or every fourth week. The number of samples and thesampling frequency were calculated from the proportion of the slaughter vol-umes at each slaughterhouse in 2002. The samples were randomly chosen andtaken by meat inspection veterinarians. The plastic sampling jars were filled with200 to 300 g of fecal material and closed tightly, leaving the air space as small aspossible. The carcass surface samples from the same animals were taken beforechilling. The brisket, the inner and outer thigh, and the pelvic cavity were

* Corresponding author. Mailing address: Finnish Food Safety Au-thority, Mustialankatu 3, Helsinki FI-00790, Finland. Phone: 358 207724471. Fax: 358 2077 24350. E-mail: marjaana.hakkinen@evira.fi.

� Published ahead of print on 16 March 2007.

3232

swabbed with two gauze pads (10 cm by 10 cm) wetted with sterile 0.1% peptonewater. Both gauze pads were placed in a sterile plastic bag, the air was squeezedout, and the bag was closed tightly. All samples were sent chilled to the NationalVeterinary and Food Research Institute (currently the Finnish Food SafetyAuthority), Helsinki, Finland. The examination started in 1 to 2 days aftersampling.

Isolation and identification of campylobacter strains. The fecal samples wereexamined by enrichment. Ten grams of fecal material was weighed and put into90 ml of Bolton broth (Campylobacter Enrichment Broth, Lab 135 plus selectivesupplement X131 [LAB M, Bury, England] plus lysed horse blood) and incu-bated at 41.5°C for 24 h in a microaerobic incubator (ThermoForma [ThermoElectron Corporation, Marietta, OH]) (O2, 5%; CO2, 10%; N2, 85%). Oneloopful (10 �l) of enrichment culture was spread onto modified Campylobactercharcoal differential agar (mCCDA) plates (Campylobacter Blood Free SelectiveMedium Lab 112 plus selective supplement X112 [LAB M, Bury, England]),which were incubated in the same conditions. In addition, one loopful (10 �l) of730 fecal samples was directly cultured on mCCDA. The surface gauze sampleswere similarly enriched in 225 ml of Bolton broth and spread onto mCCDA.

Two colonies resembling campylobacters from mCCDA plates originatingfrom direct culture and enrichment procedures were subcultured onto brucellaagar (BBL, Becton Dickinson, MD) with 5% bovine whole blood treated withsodium citrate (Finnish Food Safety Authority, Helsinki, Finland). At least twoisolates from each positive sample were identified to the species level usingmicroscopical examination of motility and cell morphology, catalase and oxidasereactions, hippurate hydrolysis, and susceptibility to nalidixic acid (26). Nalidixicacid-resistant isolates were further examined for indoxyl acetate hydrolysis andsusceptibility to cephalotin (26). Hippurate-negative, indoxyl acetate-hydrolyzingisolates were examined for H2S production in triple sugar iron agar (LAB M,Bury, England) (pH 8) and for urease production to identify Campylobacterhyointestinalis strains. The isolates were stored in brucella broth supplementedwith 15% glycerol at �70°C.

Serotyping. One to four C. jejuni isolates (287 in total) from 176 fecal samplesand 21 isolates from carcass samples were serotyped using a set of 25 commercialantisera for the serotyping of heat-stable antigens (Penner) of C. jejuni by thepassive hemagglutination method (Denka Seiken Co., Ltd., Tokyo, Japan). Testswere performed, and the results were interpreted according to the manufactur-er’s instructions.

Genotyping by PFGE. A total of 330 and 70 C. jejuni isolates from 183 fecaland 33 carcass samples, respectively, were analyzed using pulsed-field gel elec-trophoresis (PFGE). The agarose plugs were prepared according to the PulseNetprotocol (www.cdc.gov/pulsenet/protocols) and stored in Tris-EDTA buffer at4°C. DNA was digested overnight at 25°C with 20 U of SmaI restriction endo-nuclease (New England Biolabs Inc., Ipswich, MA) in a final volume of 200 �lcontaining 2 �l bovine serum albumin (New England Biolabs Inc., Ipswich, MA).PFGE was performed using the CHEF-DRIII pulsed-field electrophoresis sys-tem (Bio-Rad, CA). An agarose gel (1%) was prepared in 0.5� Tris-bufferedEDTA (Sigma-Aldrich Co, Baltimore, MD). Fragments were separated by elec-trophoresis for 18 h at 6 V and 14°C with ramped pulse times from 6.8 to 35.4 s.Salmonella serotype Braenderup strain H9812 (ATCC BAA-664) was used as thefragment size marker. The gels were stained for 45 min with ethidium bromide(0.5 �g/ml) and photographed under UV illumination. Patterns that differed byat least a single band were considered to be different subtypes. Each subtype wasnamed S1, S2, etc. The criteria presented by Tenover et al. (52) were used toassess how the subtypes were related.

Determination of antimicrobial susceptibility. The MICs of ampicillin, enro-floxacin, erythromycin, gentamicin, nalidixic acid, and oxytetracycline for 187 C.jejuni isolates from 183 rectal fecal samples were determined using a commercialbroth microdilution method, VetMIC Camp (National Veterinary Institute,Uppsala, Sweden). Epidemiological cutoff values for resistance, based on MICdistributions, were used in the interpretation of results. A C. jejuni isolate wasconsidered to be resistant to a specific antimicrobial when its MIC was distinctlyhigher than those of inherently susceptible C. jejuni isolates.

Statistical analysis. The �2 test (Excel; Microsoft Corp., Redmond, WA) wasperformed to investigate the association between month and prevalences of allcampylobacters, C. jejuni, and Campylobacter hyointestinalis subsp. hyointestinalis.

RESULTS

Prevalence. Campylobacters were detected in a total of 296out of 952 (31.1%) rectal fecal samples and in 33 out of 948(3.5%) carcass surface samples. Campylobacters were detected

in 4.4% and 37.4% of the fecal samples examined both bydirect culture and by enrichment (n � 730), respectively.

C. jejuni was detected in 186 (19.5%) and Campylobacter coliin 21 (2.2%) fecal samples. Presumptive C. hyointestinalis wasisolated from 103 (10.8%) fecal samples, but the isolates fromonly 93 samples survived after storage at �70°C and all ofthese could be confirmed as C. hyointestinalis subsp. hyointes-tinalis. Two Campylobacter species were isolated from 14 sam-ples: C. jejuni and C. hyointestinalis subsp. hyointestinalis inseven and C. jejuni and C. coli in six samples. The C. coli andC. hyointestinalis subsp. hyointestinalis isolates were detectedonly after enrichment. C. jejuni was detected in 29 (3.1%), C.coli in two (0.2%), and presumptive C. hyointestinalis in two(0.2%) carcass surface samples. In three cases the isolates fromthe fecal and carcass samples from the same animal repre-sented different Campylobacter species.

Seventy percent of the animals belonged to the age group 1to 3 years. In this age group the prevalences of C. jejuni, C.hyointestinalis subsp. hyointestinalis, and C. coli were 25.6%,10.0%, and 1.9%, respectively. In the age group that includedanimals between 3 and 7 years, which represented 25% of theanimals, the prevalences were 4.0%, 12.3%, and 3.1%, respec-tively.

The sampled animals were traced to 747 farms: 411 (43.2%)samples originated from 284 beef cattle farms and 541 (56.8%)samples from 463 dairy cattle farms (Table 1). The proportionof campylobacter-positive beef cattle farms was higher thanthat of dairy cattle farms. Campylobacter isolates originatedfrom all of the 12 abattoirs and from 255 farms. More than oneanimal (two to five) per farm was sampled on 112 occasions.Animals from 19 farms were all campylobacter positive at thesame sampling. In four cases, two Campylobacter species weredetected in animals from the same farm. Positive and negativeanimals were detected from 36 farms on the same samplingoccasion. Animals from 32 farms were sampled twice. Bothsamples from six farms were positive, and from 10 farms one ofthe samples was positive. Samples from 15 farms were campy-lobacter negative in both samplings. Two or more campy-lobacter-positive animals were detected from 33 farms eitherat the same sampling or on different occasions.

Monthly distribution. The monthly distribution of campy-lobacter-positive fecal samples is presented in Fig. 1. A slightlyincreasing trend can be seen in the overall prevalence ofcampylobacters towards the end of summer and late autumn.The prevalence of C. jejuni was highest in August and lowest inDecember. C. hyointestinalis subsp. hyointestinalis was mostfrequently isolated in November, and the lowest prevalencewas detected in April. A statistical association was observedbetween month and the overall prevalence of campylobacters,

TABLE 1. Distribution of campylobacter-positive animals betweenbeef and dairy cattle farms

Herd type No. offarms

No. ofpositive farms

%Positivefarms

No. of positiveanimals

Beef cattle 284 122 42.7 154Dairy cattle 463 133 28.7 142

Total 747 255 34.0 296

VOL. 73, 2007 PREVALENCE OF CAMPYLOBACTER SPP. IN CATTLE IN FINLAND 3233

but not C. jejuni (P � 0.05, df � 11). Also the prevalence of C.hyointestinalis subsp. hyointestinalis was statistically connectedwith month (P � 0.01, df � 11).

Serotyping. Seventeen different serotypes were detectedamong the fecal C. jejuni isolates using the commercial sero-typing kit that was employed in this study. Untypeable isolateswere obtained from 22 samples (12.5%). The predominantPenner serotypes of C. jejuni that were detected in the fecalsamples were Pen2 and Pen4-complex, which were isolatedfrom 52% of the fecal samples. In the 79 samples from whichtwo or three isolates were serotyped, the same serotype wasdetected in 28 samples, two serotypes were detected in ninesamples, and three were detected in one sample. Both typeableand untypeable isolates were obtained from the same sampleon 17 occasions. Two untypeable isolates were obtained fromfour fecal samples. In carcass samples Pen2 was detected in 11and Pen4-complex was detected in 5 out of 21 isolates.

PFGE. Fifty-six different C. jejuni subtypes were identifiedby PFGE among 330 isolates from the fecal samples. The 10most prevalent subtypes isolated from 103 fecal samples cov-ered 56.3% of C. jejuni-positive samples (Table 2). C. jejuniisolates from 164 animals (89.6%) were assigned to 21 SmaIsubtypes. Unique subtypes were isolated from 30 animals(16.4%). DNA from five isolates was not digestible with SmaI.Multiple isolates were genotyped from 106 fecal samples. Twounrelated SmaI subtypes were observed in six of them. Twoand three different but possibly related isolates were observedin two fecal samples.

When several animals from one farm were sampled at thesame time, C. jejuni isolates from animals originating from thesame farm represented indistinguishable SmaI profiles oneight occasions. Closely related subtypes were identified onone occasion and possibly related subtypes on two occasions.Unrelated genotypes were observed in five cases. On the sixoccasions when positive samples were obtained from the farmsthat were sampled twice, the C. jejuni isolates representedunrelated subtypes.

The C. jejuni isolates from carcass surface samples repre-

sented 20 different SmaI subtypes. The most frequently iso-lated subtypes were S1 and S20, which were each detected infour carcasses. Subtypes S9 and S26 were observed in threecarcasses. Eleven subtypes were detected only once. Two dif-ferent SmaI subtypes were isolated from two carcasses. In onecase the isolates were possibly related, and in the other casethey were unrelated.

On 12 occasions indistinguishable C. jejuni PFGE types weredetected in the fecal and carcass samples from the same ani-mal. Indistinguishable subtypes isolated from one animal’s fe-cal sample were detected in another animal’s carcass sample atsix samplings. In eight cases different C. jejuni subtypes wereobtained from fecal and carcass surface samples on the samesampling occasion.

Sero-/PFGE types. Twenty-three different PFGE subtypeswere observed among C. jejuni isolates classified as Pen2. Thelargest group, Pen2/S1, comprised isolates from 19 animals.Isolates belonging to Pen4-complex were split up into 13PFGE subtypes. The most common was Pen4-complex/S2,which was isolated from 10 animals. The predominant sero-

FIG. 1. Monthly distribution of campylobacters in the fecal samples.

TABLE 2. Most prevalent PFGE types of C. jejuni in fecal andcarcass samples

PFGEtype

No. of fecalsamples

% of positivefecal samples

No. ofcarcasses

% of positivecarcass samples

S1 20 10.9 4 12.1S2 14 7.7 2 6.1S3 14 7.7 0 0.0S5 10 5.5 1 3.0S6 5 2.7 2 6.1S7 10 5.5 1 3.0S9 5 2.7 3 9.1S10 6 3.3 1 3.0S11 10 5.5 2 6.1S13 7 3.8 0 0.0S14 7 3.8 1 3.0S20 0 0.0 4 12.1S26 2 1.1 3 9.1

3234 HAKKINEN ET AL. APPL. ENVIRON. MICROBIOL.

types/PFGE types are represented in Table 3. Pen2/S1 was alsothe most common type among isolates from carcass samplescomprising isolates from four carcasses. Pen2/S9, Pen2/S34,Pen4-complex/S2, and Pen1,44/S26 were all detected in twocarcasses.

Antimicrobial susceptibility of C. jejuni. Of the 187 C. jejuniisolates that were examined for antimicrobial susceptibility, 16(9%) were resistant to at least one of the antimicrobials tested(Table 4). Resistance to nalidixic acid was most commonlydetected. Six of the 11 nalidixic acid-resistant isolates were alsoresistant to enrofloxacin. No multiresistance was observedamong the isolates.

DISCUSSION

The prevalence of Campylobacter spp. in Finnish cattle atslaughter varied monthly between 18.8% and 44.1% duringthis 1-year study. In several studies performed in other coun-tries prevalences of between 7% and 100% at slaughter havebeen reported (2, 5, 14, 37, 42, 49). Due to the different studydesigns regarding various sampling methods and materials,detection methods, etc., the results are not always comparable.The sampling for our survey was carried out in 12 out of 15Finnish slaughterhouses, which covered 98% of the cattleslaughtered in Finland in 2003. The prevalence of campy-lobacters in cattle was 4.4% by direct culture and 37.3% byenrichment from the same 730 rectal fecal samples, whichsuggests that the overall level of campylobacters in the intes-tinal contents of cattle in Finland was low. This result is inaccordance with the reported average most probable numbervalues between 69/g and 6.1 � 102/g in the fecal samples of

dairy cattle from other studies (36, 49). Higher numbers ofcells have been obtained using real-time PCR for the quanti-fication of campylobacters (25).

The predominance of C. jejuni over other Campylobacterspecies has been reported in cattle by Nielsen et al. (37) and byAcik and Cetinkaya (2) and many other studies, whereas C.hyointestinalis was the species that was most frequently isolatedfrom cattle at slaughter in the surveys by Grau (17) andPezzotti et al. (45). In the present study, C. jejuni was the mostcommon species in young animals, while in the older age groupC. hyointestinalis subsp. hyointestinalis was most frequently de-tected. A similar distribution of the Campylobacter speciesamong young and adult cattle was reported by Giacoboni et al.(15). In addition to the age of the animals, the choice ofmethod can also influence the diversity of the Campylobacterspecies detected from the samples. In our study, no C. coli orC. hyointestinalis subsp. hyointestinalis isolates were obtainedby direct culture. The actual prevalence of C. hyointestinalissubsp. hyointestinalis in Finnish cattle is probably even higherthan observed in this study, where the culture medium andgrowth conditions were optimized for the selection of the ther-mophilic Campylobacter species. These cultivation methodsalso exclude more fastidious species like Campylobacter lanie-nae, which proved to be the most prevalent Campylobacterspecies in beef cattle in the studies by Inglis et al. (24, 25), whoemployed PCR methods for detection.

Significant seasonal variation in the numbers of thermophiliccampylobacters in dairy cattle herds but not in beef cattle hasbeen reported by Stanley et al. (49). No evidence of the influ-ence of climatic factors was observed, and the authors sug-gested that increased fecal excretion of campylobacters wasdue to hormonal factors or changes in the water supply anddiet. In our study the overall patterns of monthly distributionof campylobacters in beef and dairy cattle were similar (datanot shown). C. jejuni and C. hyointestinalis subsp. hyointestina-lis, however, showed slightly different monthly patterns. Theincreasing prevalence of C. jejuni towards the end of the sum-mer, although not significant, may reflect the continuous chal-lenge during the grazing period (June to September) originat-ing from environmental sources such as drinking water (20, 23)in contrast to the winter period, when the cattle are kept insidein Finland and given tap water to drink. No obvious reasoncould be found for C. hyointestinalis subsp. hyointestinalisreaching its highest level in November. In the Nordic countries,a seasonal peak in reported human campylobacter infections as

TABLE 3. Predominant combined sero-/PFGE types ofC. jejuni isolates from fecal samples

Penner serotype SmaIsubtype

No. of positivesamples

% of all positivesamples

2 S1 19 10.82 S5 10 5.72 S11 7 4.02 S18 5 2.84-complex S2 10 5.74-complex S3 7 4.012 S7 8 4.51,44 S6 5 2.8

Total 71 40.3

TABLE 4. Distribution of MICs among C. jejuni isolates

Antimicrobial% Resistant

isolates(95% CIa)

Breakpoint forresistance(mg/liter)

Range of dilutionstested (mg/liter)

% of isolates with MICb (mg/liter):

�0.03 0.06 0.12 0.25 0.5 1 2 4 8 16 32 64 128 256

Ampicillin 1.6 (0.3–4.6) 16 0.5–64 5.9 7.0 41.2 40.1 3.2 1.1 0.5 1.1Enrofloxacin 3.2 (1.2–6.9) 0.5 0.03–4 1.1 8.0 49.2 33.7 4.8 1.1 0.0 1.6 0.5c

Erythromycin 0 (0.0–2.0) 8 0.12–16 1.1 1.6 22.5 51.9 20.9 2.1Gentamicin 0 (0.0–2.0) 4 0.25–8 3.2 54.0 42.2 0.5Nalidixic acid 5.9 (3.0–10.3) 16 1–128 1.6 15.5 61.0 16.0 3.7 0.0 0.5 1.6c

Tetracycline 1.1 (0.1–3.8) 2 0.25–32 92.0 5.9 0.5 0.5 0.5 0.5

a CI, confidence interval.b MICs equal to or lower than the lowest concentration tested are given as the lowest concentration.c MIC greater than the highest concentration in the range of dilutions tested.

VOL. 73, 2007 PREVALENCE OF CAMPYLOBACTER SPP. IN CATTLE IN FINLAND 3235

well as in the number of campylobacter-positive broiler flockshas consistently been observed in late summer (22, 35a, 40).

The proportion of campylobacter-positive cattle farms waslow, on average 34%, in the present study. In a Danish study C.jejuni was present on 83% of dairy farms (36). The low numberof samples per farm may explain the low percentage of positivefarms in our study. Beef cattle were more frequently colonizedby campylobacters than dairy cattle. A similar observation wasreported by Beach et al. (5) and Grau (17), who suggested thatthe diet and high animal density of lot-fed cattle encouragedthe intestinal colonization and spread of campylobacters. Thevariation in the colonization of beef and dairy cattle observedin our study may, however, reflect the age of the animals ratherthan the type of the herd, because most of the beef cattle wereslaughtered at the age of 1 to 2 years, whereas most dairy cattlewere slaughtered between the ages of 3 and 7 years. A higherprevalence of campylobacters in young animals has been ob-served in other studies (17, 36).

The predominant Penner serotypes of C. jejuni observed inthis study (Pen2, Pen4-complex, Pen1,44, and Pen12) were alsocommon in the human infections originating in Finland duringthe period July to September 1999 (53), but the most prevalentserotype in the human cases originating in Finland, Pen6,7, wasrarely observed in cattle. The Pen4-complex and Pen12 sero-types have also been reported in Finnish poultry, althoughPen6,7 was predominant (44). The percentage of fecal samplesthat yielded untypeable isolates was 12.5%, which is in accor-dance with the results from other studies, where commercialantisera from the same manufacturer were used for serotyping(44, 46). In our study, Pen2 was most frequently isolated fromcattle in June (data not shown), whereas Vierikko et al. (53)reported a peak in the occurrence of the same serotype inhumans in Finland in August. It would be interesting to findout whether these two peaks really do follow each other, sug-gesting that cattle may play a role in human infections. Thesedata, however, originate from different years, and the annualvariation cannot be excluded. The second most prevalent se-rotype from bovine samples, Pen4-complex, showed a differentseasonal pattern with a peak in September regarding cattle, butit was at its highest in humans in August (53). These twoserotypes were also the most commonly detected in dairy cattlein other studies (9, 27, 36, 37), which suggests that they may beparticularly adapted to colonizing the bovine gut. Cocoloniza-tion by two serotypes in 8% of animals was reported by Nielsen(36). In the present study concurrent colonization by two C.jejuni serotypes was observed in 39% of animals from whichtwo or more isolates were serotyped, assuming that the un-typeable isolates represent different serotypes from the iden-tified serotypes in the same sample.

Genotyping by PFGE revealed a high degree of diversityamong the bovine C. jejuni isolates. This has been seen in otherstudies with other typing methods as well (2, 6, 48) and also inregard to C. jejuni isolates from chickens, sheep, turkeys, wa-ter, and human cases (9, 13, 38). A wide variation of SmaIsubtypes could be observed among the isolates representingthe most common serotype, Pen2. This has been found previ-ously in the Danish study by Nielsen (36). Genomic instability,which enables the adaptation of the organism to variable en-vironmental conditions (19, 54), has been given as the expla-nation for the diversity. However, significant genomic stability

and clonal lineages of certain C. jejuni serotypes from a varietyof hosts and geographic areas have been reported (29, 31).Despite the small number of isolates from each sample, morethan one SmaI subtype was identified from 12% of the fecalsamples from which multiple isolates were genotyped. Thepresence of unrelated subtypes in the samples suggests thatthere may have been several sources of campylobacters on thefarm.

Although the sampling was planned only for investigatingthe situation at slaughter, the tracing of the animals to theirfarm also made some considerations possible at the farm levelas well. When more than one animal was sampled at a time perfarm, most commonly undistinguished or closely related sub-types were isolated from C. jejuni-positive samples. The coex-istence of two or three unrelated C. jejuni subtypes or differentCampylobacter species in the samples from a farm was ob-served in few cases. These observations might suggest animal-to-animal transmission or one or a small number of commonsources of contamination (6, 36). Closely related isolates wererarely detected on a farm, which may reflect either the geneticinstability of the strains or the temporary colonization of theanimals. An indication of the latter may also be the detectionof campylobacter-positive and -negative samples from thesame farms at the same sampling. The observation that only aportion of the animals are simultaneously colonized is possiblydue to the intermittent excretion of campylobacters or lownumbers of campylobacters in the samples (36).

Campylobacters were not detected in almost half the caseswhen animals from the same farms were sampled twice. Whenthis and the low prevalence of campylobacters in cattle atslaughter in this study are taken into consideration, it may bepossible that cattle farms which are always campylobacter neg-ative do exist. On the other hand, it may also reflect lownumbers of campylobacters in the fecal samples.

Campylobacter contamination rates of 0 to 25% of carcassesbefore chilling and 3% after chilling have been reported inother studies (5, 17, 34). Due to the sensitivity of campy-lobacters to oxygen and drying, air chilling reduces the con-tamination of the carcasses (16, 17, 43). In the present surveythe contamination level of carcasses was low (3.5%) beforechilling, which may reflect the low number of campylobactersin cattle feces but probably indicates good slaughter hygiene aswell and suggests that contamination of beef at the retail levelis very low. Obviously, during the slaughter process cross-con-tamination can originate from the feces of the same animal ordifferent animals through the slaughterhouse environment orequipment. The C. jejuni serotypes most frequently isolatedfrom carcasses were the same as those isolated from the feces.Comparison of the PFGE subtypes from fecal and carcasssamples revealed, however, that some subtypes commonly de-tected in fecal samples were not isolated from carcasses. Thismay indicate variation between subtypes regarding tolerance tooxygen and drying. One of the most common subtypes in car-cass samples was not, however, isolated in feces. It may bepossible that subtypes exist which are poor competitors in theintestines but can survive in the conditions on the surface ofthe carcass.

The overall prevalence of antimicrobial resistance amongbovine fecal C. jejuni isolates was low. A small proportion of C.jejuni isolates were resistant to ampicillin, tetracycline, and

3236 HAKKINEN ET AL. APPL. ENVIRON. MICROBIOL.

enrofloxacin. Aminopenicillins, fluoroquinolones, and tetracy-clines are used in the treatment of bovine infectious diseases inFinland. No resistance to erythromycin was detected, althoughmacrolides are used in the treatment of bovine infections.Resistance to nalidixic acid was almost twice as common asresistance to enrofloxacin. Similar findings on the resistance ofbovine campylobacters to quinolones have been described byAarestrup et al. (1) and Englen et al. (11). Comparison withresistance data from other countries is complicated by varia-tions in the methodologies and breakpoints that are used toclassify the isolates as resistant. Breakpoints recommended forEnterobacteriaceae by CLSI (formerly NCCLS) have usuallybeen applied in previous studies, as no internationally agreedclinical or epidemiological breakpoints for antimicrobial resis-tance of campylobacters have been available. In a recent pub-lication by CLSI (8) criteria are presented for erythromycin(�32 �g/ml), ciprofloxacin (�4 �g/ml), and tetracycline (�16�g/ml). Interpretation of MICs according to these criteriawould have yielded less than 5% total resistance among thebovine C. jejuni isolates in the present study, which is substan-tially lower than that reported from other European countriesand the United States (1, 4, 7, 11).

In conclusion, the prevalence of campylobacters in Finnishcattle at slaughter was low and carcass contamination was rarein this survey, indicating that Finnish beef can be considered asa minor source of campylobacters for consumers. The antimi-crobial resistance level among bovine C. jejuni isolates was alsolow, and multiresistance was not detected, which may be ex-plained by the prudent use of antimicrobial agents for animals.However, the common occurrence of serotypes Pen2 andPen4-complex in cattle indicates that there may be an indirectassociation with human infections.

ACKNOWLEDGMENTS

This work has been supported by the Walter Ehrstrom Foundationand the Finnish Veterinary Science Foundation.

We thank Kirsi Eklund, Maaret Hypponen, Mira Kankare, LeaNygård, and Kaija Pajunen for their excellent technical assistance andLeila Rantala for assistance in interpreting the PFGE profiles.

REFERENCES

1. Aarestrup, F. M., E. M. Nielsen, M. Madsen, and J. Engberg. 1997. Anti-microbial susceptibility patterns of thermophilic Campylobacter spp. fromhumans, pigs, cattle, and broilers in Denmark. Antimicrob. Agents Che-mother. 41:2244–2250.

2. Acik, M. N., and B. Cetinkaya. 2005. The heterogeneity of Campylobacterjejuni and Campylobacter coli strains isolated from healthy cattle. Lett. Appl.Microbiol. 41:397–403.

3. Anderson, S. A., R. W. Yeaton Woo, and L. M. Crawford. 2001. Risk assess-ment of the impact on human health of resistant Campylobacter jejuni fromfluoroquinolone use in beef cattle. Food Control 12:13–25.

4. Bae, W., K. N. Kaya, D. D. Hancock, D. R. Call, Y. H. Park, and T. E. Besser.2005. Prevalence and antimicrobial resistance of thermophilic Campylobacterspp. from cattle farms in Washington State. Appl. Environ. Microbiol. 71:169–174.

5. Beach, J. C., E. A. Murano, and G. R. Acuff. 2002. Prevalence of Salmonellaand Campylobacter in beef cattle from transport to slaughter. J. Food Prot.65:1687–1693.

6. Besser, T. E., J. T. LeJeune, D. H. Rice, J. Berg, R. P. Stilborn, K. Kaya, W.Bae, and D. D. Hancock. 2005. Increasing prevalence of Campylobacter jejuniin feedlot cattle through the feeding period. Appl. Environ. Microbiol. 71:5752–5758.

7. Bywater, R., H. Deluyker, E. Deroover, A. de Jong, H. Marion, M. McConville,T. Rowan, T. Shryock, D. Shuster, V. Thomas, M. Valle, and J. Walters. 2004.A European survey of antimicrobial susceptibility among zoonotic and com-mensal bacteria isolated from food-producing animals. J. Antimicrob. Che-mother. 54:744–754.

8. Clinical and Laboratory Standards Institute. 2006. Methods for antimicro-

bial dilution and disk susceptibility testing of infrequently isolated or fastid-ious bacteria; approved guideline M45-A. Clinical and Laboratory StandardsInstitute, Wayne, PA.

9. Devane, M. L., C. Nicol, A. Ball, J. D. Klena, P. Scholes, J. A. Hudson, M. G.Baker, B. J. Gilpin, N. Garret, and M. G. Savill. 2005. The occurrence ofCampylobacter subtypes in environmental reservoirs and potential transmis-sion routes. J. Appl. Microbiol. 98:980–990.

10. Endtz, H. P., G. J. Ruijs, B. van Klingeren, W. H. Jansen, T. van der Reyden,and R. P. Mouton. 1991. Quinolone resistance in campylobacter isolatesfrom man and poultry following the introduction of fluoroquinolones inveterinary medicine. J. Antimicrob. Chemother. 27:199–208.

11. Englen, M. D., P. J. Fedorka-Cray, S. R. Ladely, and D. A. Dargatz. 2005.Antimicrobial resistance patterns of Campylobacter from feedlot cattle.J. Appl. Microbiol. 99:285–291.

12. Evans, M. R., R. J. Roberts, C. D. Ribeiro, D. Gardner, and D. Kembrey.1996. A milk-borne campylobacter outbreak following an educational farmvisit. Epidemiol. Infect. 117:457–462.

13. Fitzgerald, C., K. Stanley, S. Andrew, and K. Jones. 2001. Use of pulsed-fieldgel electrophoresis and flagellin gene typing in identifying clonal groups ofCampylobacter jejuni and Campylobacter coli in farm and clinical environ-ments. Appl. Environ. Microbiol. 67:1429–1436.

14. Garcia, M. M., H. Lior, R. B. Stewart, G. M. Ruckerbauer, J. R. R. Trudel,and A. Skljarevski. 1985. Isolation, characterization, and serotyping ofCampylobacter jejuni and Campylobacter coli from slaughter cattle. Appl.Environ. Microbiol. 49:667–672.

15. Giacoboni, G. I., K. Itoh, K. Hirayama, E. Takahashi, and T. Mitsuoka.1993. Comparison of faecal Campylobacter in calves and cattle of differentages and areas in Japan. J. Vet. Med. Sci. 55:555–559.

16. Gill, C. O., and L. M. Harris. 1982. Contamination of red-meat carcasses byCampylobacter fetus subsp. jejuni. Appl. Environ. Microbiol. 43:977–980.

17. Grau, F. H. 1988. Campylobacter jejuni and Campylobacter hyointestinalis inthe intestinal tract and on the carcasses of calves and cattle. J. Food Prot.51:857–861.

18. Gupta, A., J. M. Nelson, T. J. Barrett, R. V. Tauxe, S. P. Rossiter, C. R.Friedman, K. W. Joyce, K. E. Smith, T. F. Jones, M. A. Hawkins, B. Shiferaw,J. L. Beebe, D. J. Vugia, T. Rabatsky-Ehr, J. A. Benson, T. P. Root, and J.Angulo. 2004. Antimicrobial resistance among Campylobacter strains, UnitedStates, 1997–2001. Emerg. Infect. Dis. 10:1102–1109.

19. Hanninen, M.-L., M. Hakkinen, and H. Rautelin. 1999. Stability of relatedhuman and chicken Campylobacter jejuni genotypes after passage throughchick intestine studied by pulsed-field gel electrophoresis. Appl. Environ.Microbiol. 65:2272–2275.

20. Hanninen, M.-L., M. Niskanen, and L. Korhonen. 1998. Water as a reservoirfor Campylobacter jejuni infection in cows studied by serotyping and pulsed-field gel electrophoresis (PFGE). J. Vet. Med. B 45:37–42.

21. Hanninen, M.-L., P. Perko-Makela, A. Pitkala, and H. Rautelin. 2000. Athree-year study of Campylobacter jejuni genotypes in humans with domes-tically acquired infections and in chicken samples from the Helsinki area.J. Clin. Microbiol. 38:1998–2000.

22. Hofshagen, M., and H. Kruse. 2005. Reduction in flock prevalence ofCampylobacter spp. in broilers in Norway after implementation of an actionplan. J. Food Prot. 68:2220–2223.

23. Humphrey, T. J., and B. Beckett. 1987. Campylobacter jejuni in dairy cowsand raw milk. Epidemiol. Infect. 98:263–269.

24. Inglis, G. D., L. D. Kalischuk, and H. W. Busz. 2003. A survey of Campy-lobacter species shed in faeces of beef cattle using polymerase chain reaction.Can. J. Microbiol. 49:655–661.

25. Inglis, G. D., L. D. Kalischuk, and H. W. Busz. 2004. Chronic shedding ofCampylobacter species in beef cattle. J. Appl. Microbiol. 97:410–420.

26. International Organization for Standardization. 2006. Microbiology of foodand animal feeding stuffs—horizontal method for detection and enumera-tion of Campylobacter spp. Part 1: detection method. International Organi-zation for Standardization, Geneva, Switzerland.

27. Ishihara, K., T. Yamamoto, S. Satake, S. Takayama, S. Kubota, H. Negishi,A. Kojima, T. Asai, T. Sawada, T. Takahashi, and Y. Tamura. 2006. Com-parison of Campylobacter isolated from humans and food-producing animalsin Japan. J. Appl. Microbiol. 100:153–160.

28. Karenlampi, R., H. Rautelin, M. Hakkinen, and M.-L. Hanninen. 2003.Temporal and geographical distribution and overlap of Penner heat-stableserotypes and pulsed-field electrophoresis genotypes of Campylobacter jejuniisolates collected from humans and chickens in Finland during a seasonalpeak. J. Clin. Microbiol. 41:4870–4872.

29. Laturnus, C., J. Jores, I. Moser, P. Schwerk, and L. H. Wieler. 2005. Long-term clonal lineages within Campylobacter jejuni O:2 strains from differentgeographical regions and hosts. Int. J. Med. Microbiol. 294:521–524.

30. Lehner, A., C. Schneck, G. Feierl, P. Pless, A. Deutz, E. Brandl, and M.Wagner. 2000. Epidemiologic application of pulsed-field gel electrophoresisto an outbreak of Campylobacter jejuni in an Austrian youth centre. Epide-miol. Infect. 125:13–16.

31. Manning, G., B. Duim, T. Wassenaar, J. A. Wagenaar, A. Ridley, and D. G.Newell. 2001. Evidence for a genetically stable strain of Campylobacter jejuni.Appl. Environ. Microbiol. 67:1185–1189.

VOL. 73, 2007 PREVALENCE OF CAMPYLOBACTER SPP. IN CATTLE IN FINLAND 3237

32. Meldrum, R. J., J. K. Griffiths, R. M. M. Smith, and M. R. Evans. 2005. Theseasonality of human campylobacter infection and Campylobacter isolatesfrom fresh, retail chicken in Wales. Epidemiol. Infect. 133:49–52.

33. Michaud, S., S. Menard, and R. D. Arbeit. 2004. Campylobacteriosis, East-ern Townships, Quebec. Emerg. Infect. Dis. 10:1844–1847.

34. Minihan, D., P. Whyte, M. O’Mahony, S. Fanning, K. McGill, and J. D.Collins. 2004. Campylobacter spp. in Irish feedlot cattle: a longitudinal studyinvolving pre-harvest and harvest phases of the food chain. J. Vet. Med. B51:28–33.

35. National Public Health Institute. 2005. Infectious diseases in Finland 1995–2004. National Public Health Institute, Helsinki, Finland.

35a.National Veterinary Institute. 2004. Trends and sources of zoonoses andzoonotic agents in humans, foodstuffs, animals and feedingstuffs includinginformation on foodborne outbreaks and antimicrobial resistance in zoo-notic agents in 2004. National Veterinary Institute, Uppsala, Sweden. http://www.sva.se/pdf/zoonosis/zoonosrapport_Sverige_2004.pdf.

36. Nielsen, E. M. 2002. Occurrence and strain diversity of themophilic campy-lobacters in cattle of different age groups in dairy herds. Lett. Appl. Micro-biol. 35:85–89.

37. Nielsen, E. M., J. Engberg, and M. Madsen. 1997. Distribution of serotypesof Campylobacter jejuni and C. coli from Danish patients, poultry and swine.FEMS Immunol. Med. Microbiol. 19:47–56.

38. Nielsen, E. M., V. Fussing, J. Engberg, N. L. Nielsen, and J. Neimann. 2006.Most Campylobacter subtypes from sporadic infections can be found in retailpoultry products and food animals. Epidemiol. Infect. 134:758–767.

39. Nygård, K., Y. Andersson, J. A. Røttingen, Å. Svensson, J. Lindback, T.Kistemann, and J. Giesecke. 2004. Association between environmental riskfactors and campylobacter infections in Sweden. Epidemiol. Infect. 132:317–325.

40. Nylen, G., F. Dunstan, S. R. Palmer, Y. Andersson, F. Bager, J. Cowden, G.Feierl, Y. Galloway, G. Kapperud, F. Megraud, K. Molbak, L. R. Petersen,and P. Ruutu. 2002. The seasonal distribution of campylobacter infection innine European countries and New Zealand. Epidemiol. Infect. 128:383–390.

41. Olah, P., J. S. Sherwood, L. M. Elijah, M. R. Dockter, C. Doetkott, Z. Miller,and C. M. Logue. 2004. Comparison of antimicrobial resistance in Salmo-nella and Campylobacter isolated from turkeys in the Midwest USA. FoodMicrobiol. 21:779–789.

42. Ono, K., and K. Yamamoto. 1999. Contamination of meat with Campy-lobacter jejuni in Saitama. Int. J. Food Microbiol. 47:211–219.

43. Oosterom, J., G. J. A. de Wilde, E. de Boer, L. H. de Blaauw, and H. Karman.1983. Survival of Campylobacter during poultry processing and pig slaugh-tering. J. Food Prot. 46:702–706.

44. Perko-Makela, P., M. Hakkinen, T. Honkanen-Buzalski, and M.-L. Hanninen.2002. Prevalence of campylobacters in chicken flocks during the summer of 1999in Finland. Epidemiol. Infect. 129:187–192.

45. Pezzotti, G., A. Serafin, I. Luzzi, R. Mioni, M. Milan, and R. Perin. 2003.Occurrence and resistance to antibiotics of Campylobacter jejuni and Campy-lobacter coli in animals and meat in northeastern Italy. Int. J. Food Micro-biol. 82:281–287.

46. Rautelin, H., and M.-L. Hanninen. 1999. Comparison of a commercial testfor serotyping heat-stable antigens of Campylobacter jejuni with genotypingby pulsed-field gel electrophoresis. J. Med. Microbiol. 48:617–621.

47. Schildt, M., S. Savolainen, and M.-L. Hanninen. 2006. Long-lasting Campy-lobacter jejuni contamination of milk associated with gastrointestinal illnessin a farming family. Epidemiol. Infect. 134:401–405.

48. Siemer, B. L., C. S. Harrington, E. M. Nielsen, B. Borck, N. L. Nielsen, J.Engberg, and S. L. W. On. 2004. Genetic relatedness among Campylobacterjejuni serotyped isolates of diverse origin as determined by numerical anal-ysis of amplified fragment length polymorphism (AFLP) profiles. J. Appl.Microbiol. 96:795–802.

49. Stanley, K. N., J. S. Wallace, J. E. Currie, P. J. Diggle, and K. Jones. 1998.The seasonal variation of thermophilic campylobacters in beef cattle, dairycattle and calves. J. Appl. Microbiol. 85:472–480.

50. Stern, N. J., M. P. Hernandez, L. Blankenship, K. E. Deibel, S. Doores, M. P.Doyle, H. Ng, M. D. Pierson, J. N. Sofos, W. H. Sveum, and D. C. Westhoff.1985. Prevalence and distribution of Campylobacter jejuni and Campylobactercoli in retail meats. J. Food Prot. 48:595–599.

51. Studahl, A., and Y. Andersson. 2000. Risk factors for indigenous campy-lobacter infection: a Swedish case control study. Epidemiol. Infect. 125:269–275.

52. Tenover, F. C., R. D. Arbeit, R. V. Goering, P. A. Mickelsen, B. E. Murray,D. H. Persing, and B. Swaminathan. 1995. Interpreting chromosomal DNArestriction patterns produced by pulsed-field gel electrophoresis: criteria forbacterial strain typing. J. Clin. Microbiol. 33:2233–2239.

53. Vierikko, A., M.-L. Hanninen, A. Siitonen, P. Ruutu, and H. Rautelin. 2004.Domestically acquired Campylobacter infections in Finland. Emerg. Infect.Dis. 10:127–130.

54. Wassenaar, T. M., B. Geilhausen, and D. G. Newell. 1998. Evidence ofgenomic instability in Campylobacter jejuni isolated from poultry. Appl. En-viron. Microbiol. 64:1816–1821.

55. Whyte, P., K. McGill, D. Cowley, R. H. Madden, L. Moran, P. Scates, C.Carroll, A. O’Leary, S. Fanning, J. D. Collins, E. McNamara, J. E. Moore,and M. Cormican. 2004. Occurrence of Campylobacter in retail foods inIreland. Int. J. Food Microbiol. 95:111–118.

55a.World Health Organization. 2001. The increasing incidence of humancampylobacteriosis. Report and proceedings of a WHO consultation ofexperts. Copenhagen, Denmark, 21 to 25 November 2000. World HealthOrganization, Geneva, Switzerland. http://whqlibdoc.who.int/hq/2001/who_cds_csr_aph_2001.7.pdf.

3238 HAKKINEN ET AL. APPL. ENVIRON. MICROBIOL.

II

ORIGINAL ARTICLE

Shedding of Campylobacter spp. in Finnish cattleon dairy farmsM. Hakkinen1 and M.-L. Hanninen2

1 Research Department, Finnish Food Safety Authority, Helsinki, Finland

2 Veterinary Faculty, Department of Food and Environmental Hygiene, University of Helsinki, Helsinki, Finland

Introduction

The number of reported cases of campylobacteriosis, the

most common reported bacterial enteric infection in

humans, has steadily increased in most countries. Several

studies have identified poultry meat as the most important

food item associated with sporadic cases of campylobacter

infection (Studahl and Andersson 2000; Neimann et al.

2003; Wingstrand et al. 2006; Gormley et al. 2008). This

rising trend in human campylobacteriosis is also evident in

the Nordic countries, where the prevalence of campylobact-

ers in poultry flocks is low (Anon. 2007a), thus suggesting

other possible sources of these organisms.

Among other food production animals, cattle are iden-

tified as common carriers of Campylobacter jejuni (Besser

et al. 2005; Devane et al. 2005; Kwan et al. 2008), but the

occurrences of campylobacters on cattle carcasses and in

beef are low (Minihan et al. 2004; Whyte et al. 2004;

Hakkinen et al. 2007). Instead, unpasteurized milk has

emerged as a risk factor for human campylobacteriosis in

epidemiological studies (Studahl and Andersson 2000;

Neimann et al. 2003) and has caused numerous outbreaks

(Evans et al. 1996; Lehner et al. 2000; Schildt et al. 2005).

In addition, indirect exposure to cattle faeces through

environmental contamination is considered a high risk to

humans (Minihan et al. 2004; Devane et al. 2005; Garrett

et al. 2007). In a wide water-borne outbreak in Canada,

one cattle farm was implicated in the contamination by

C. jejuni of the municipal drinking water supply (Clark

et al. 2003).

Longitudinal studies on the persistence of campylobac-

ter colonization among beef cattle have been performed

Keywords

Arcobacter, Campylobacter jejuni, dairy cattle,

PFGE.

Correspondence

Marjaana Hakkinen, Research Department,

Finnish Food Safety Authority, Mustialankatu

3, FI 00790, Finland.

E-mail: marjaana.hakkinen@evira.fi

2008 ⁄ 1681: received 1 October 2008, revised

2 January 2009 and accepted 21 January

2009

doi:10.1111/j.1365-2672.2009.04269.x

Abstract

Aims: The aim of this study was to determine variation of prevalence through-

out a year, colonization levels and genotypes of Campylobacter jejuni in Finnish

dairy cattle herds.

Methods and Results: Faecal samples and tank milk samples from three dairy

cattle herds were taken five times, and swab samples from drinking troughs

once during a 1-year sampling period. The samples were enriched in Bolton

broth and subsequently spread on mCCDA. Isolates were then subtyped by

pulsed-field gel electrophoresis using SmaI. Campylobacter jejuni was detected

in 169 of the 340 faecal samples and in one drinking trough sample.

Prevalences between herds and sampling times varied widely. The faecal levels

of C. jejuni were mainly low. Between one and four SmaI subtypes were identi-

fied from each herd per sampling. Two SmaI subtypes persisted in two of the

herds throughout the study.

Conclusions: Dairy cattle can be a long-term reservoir of C. jejuni subtypes

similar to clinical isolates. Differences in the colonization potential among

C. jejuni strains as well as in the resistance to campylobacter colonization

among animals are possible.

Significance and Impact of the Study: The study provides data on contamina-

tion dynamics, colonization levels and the persistence of C. jejuni in dairy

cattle.

Journal of Applied Microbiology ISSN 1364-5072

898 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905

ª 2009 The Authors

by, e.g. Minihan et al. (2004), Besser et al. (2005) and

Kwan et al. (2008). Stanley et al. (1998) carried out a

study on seasonal fluctuation in the prevalence and num-

bers of campylobacters in cattle, including dairy herds.

There are also other interesting issues concerning dairy

herds because of their longer life span than that of beef

cattle. If permanent colonization of dairy cattle by human

pathogenic campylobacter genotypes occur, it can main-

tain the environmental load of pathogenic strains. Few

data are available on the persistence of different C. jejuni

genotypes in dairy herds.

Our study aimed to obtain data on fluctuation in intesti-

nal colonization throughout a year, colonization levels and

genotypes of C. jejuni in Finnish dairy cattle herds.

Materials and methods

Sampling

Three dairy cattle herds located 60 km apart from each

other in Southern Finland were included in the study.

Campylobacter jejuni was previously detected in pooled

faecal samples from each of the herds. The number of

animals in herds 1, 2 and 3 was 15, 20 and 90, respec-

tively. Between 17 and 33 samples of fresh, newly avoided

faeces per herd were collected from the floor on five

sampling occasions during the study: (i) after the grazing

period in November 2006, (ii) in the middle of the winter

housing period in January–February 2007, (iii) before the

new grazing period in April 2007, (iv) during the grazing

period in August 2007 and (v) after the grazing period in

November 2007. Animals recently treated with antimicro-

bials were excluded from the sampling. When possible,

the individual identification codes of animals were

included in the sampling data. Tank milk samples

(1000 ml) were also taken on each sampling occasion.

In addition, sponge swab (Medical Wire & Equipment,

Corsham, Wiltshire, UK) samples from drinking troughs

were taken during the last sampling in November 2007.

The samples were chilled and transported to the labora-

tory, and the analyses began later on the same day. The

faecal and drinking trough samples were analysed at the

Finnish Food Safety Authority Evira, Research Depart-

ment, Microbiology Unit, and the milk samples were

examined at Helsinki University, Veterinary Faculty,

Department of Food and Environmental Hygiene.

Isolation, semiquantitative detection and the

identification of campylobacters in faecal and sponge

swab samples

All the samples were analysed individually. To detect

campylobacters, 10 g of faeces were enriched in 90 ml of

Bolton broth [Campylobacter Enrichment Broth, Lab

135 + selective supplement X131 (LAB M, Bury,

England) + lysed horse blood]. In addition, a 10-fold

dilution series up to 10)6 was made using 9-ml tubes of

Bolton broth for the semiquantitative detection of campy-

lobacters (Anon. 2007b). Sponge swab samples were

enriched in 225 ml of Bolton broth. The enrichment

cultures were incubated for 24 h at 41Æ5�C and cultured

onto modified charcoal cefoperazone deoxycholate agar

[Campylobacter Blood Free Selective Medium Lab

112 + selective supplement X112 (LAB M)] as described

in Hakkinen et al. (2007).

When possible, a minimum of five typical colonies was

isolated per sample, mostly from the highest dilution

where growth was observed, but also from lower dilu-

tions, if they contained separate colonies, and especially

when colony morphology varied. Isolates were identified

at the species level according to ISO 10272-1 (Anon.

2006). To identify C. hyointestinalis ssp. hyointestinalis

strains among hippurate-negative and indoxyl acetate-

hydrolysing isolates, H2S production in TSI agar (LAB

M) (pH 8) and urease production were examined. The

isolates were stored in Brucella broth supplemented with

15% glycerol at )70�C.

Enumeration of campylobacters in milk samples

The most probable number (MPN) technique was used

to enumerate campylobacters in milk samples. Either

10 · 100 ml (November–February) or 10 · 20 ml (April–

December) of raw milk was enriched in Bolton broth

(100 ml milk + 500 ml Bolton broth or 20 ml milk + 80

ml of Bolton broth). The enrichment cultures were incu-

bated microaerobically at 37�C for 48 h and plated on

mCCDA plates which were incubated microaerobically at

37�C for 48 h. Isolates were identified according to ISO

10272-1 (Anon. 2006).

Genotyping by pulsed-field gel electrophoresis (PFGE)

A minimum of two C. jejuni isolates per sample were

analysed with PFGE using SmaI for the restriction enzyme

as described previously by Hakkinen et al. (2007). PFGE

data were analysed with Bionumerics ver. 5.10 (Applied

Maths, Kortrijk, Belgium), with 0Æ5% optimization and

1Æ0% tolerance.

Results

In total, C. jejuni was detected in 169 of the 340 faecal

samples and in one of the sponge swab samples from the

drinking troughs. No campylobacters were detected in the

milk samples. Campylobacter coli and C. hyointestinalis

M. Hakkinen and M.-L. Hanninen Campylobacter spp. in cattle

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905 899

ssp. hyointestinalis were detected in 11 (3Æ2%) and 52

(15Æ3%) of the faecal samples, respectively. The average

prevalence of C. jejuni throughout the study was 44%,

and the average monthly prevalences were 64%, 43%,

37%, 38% and 45% in November 2006, February 2007,

April 2007, August 2007 and November 2007, respec-

tively. In herds 1 and 2, the prevalence of C. jejuni was

highest in November 2006 and decreased in the winter,

whereas in herd 3, the prevalence of C. jejuni was high

(82–90%) in November 2006, January and April 2007 and

only slightly lower in August (Fig. 1). Campylobacter hyo-

intestinalis ssp. hyointestinalis was detected in samples

from herds 2 and 3. In addition, catalase- and urease-neg-

ative, H2S-producing Campylobacter sp. was detected in

herd 1 throughout the sampling period. Concurrent colo-

nization by C. jejuni and C. hyointestinalis ssp. hyointesti-

nalis occurred in 11 samples from farm 3.

The levels of C. jejuni in the faecal samples were low in

general (Table 1). In approx. 42% of the positive faecal

samples, the highest dilution in which campylobacters

were detected was 10)2. Campylobacters were detected

from the highest dilution in only four samples. In herd 3,

where the prevalence was consistently higher than in the

two other herds, high levels of C. jejuni were also detected

on all sampling occasions except in August 2007. In the

animals of herds 1 and 2, high levels were observed only

occasionally.

In total, 13 different SmaI subtypes were distinguished

among faecal C. jejuni isolates (Fig. 2). One of the sub-

types, S7, was detected in herds 1 and 2. One to four

subtypes were detected from each of the herds on each

sampling occasion (Table 2), except in August, when no

C. jejuni was detected in herd 2. Two C. jejuni SmaI

subtypes existed in herds 1 and 3 during the entire

sampling period (Fig. 3). In August 2007, a few new sub-

types emerged in both of the herds. In herd 2, only two

C. jejuni subtypes occurred throughout the entire

sampling period.

From 150 of the positive samples, two isolates per

sample were subtyped, and from 19 samples, four to six

isolates were subtyped. Two different SmaI subtypes were

detected in 3 of the 169 positive samples.

Ten animals from herd 1 were sampled on every

sampling occasion. Campylobacter jejuni was detected in

all the samples of one animal, whereas two of the animals

were campylobacter-negative on all sampling occasions

(Table 3). Three animals were campylobacter-positive

only once: two of them at a low level and one at a high

level. One SmaI subtype was consistently isolated from

each of the animals that yielded multiple positive samples,

with the exception of a previous carrier of subtype S7,

0

20

40

60

80

100

November 06 November 07 February 07 April 07 August 07

Sampling time

%

Figure 1 Prevalence of Campylobacter jejuni in three dairy cattle

herds on different sampling occasions between November 2006 and

November 2007. Herd 1; Herd 2; and Herd 3.

Table 1 Contamination levels of Campylobacter jejuni in faecal samples from three dairy cattle herds on different sampling occasions between

November 2006 and November 2007

Sample size

in enrichment

(g)

Number of Campylobacter jejuni-positive samples

Total

Herd 1 Herd 2 Herd 3

Nov

2006

Feb

2007

Apr

2007

Aug

2007

Nov

2007

Nov

2006

Feb

2007

Apr

2007

Aug

2007

Nov

2007

Nov

2006

Feb

2007

Apr

2007

Aug

2007

Nov

2007

10 3 0 0 1 2 3 1 1 0 2 8 3 8 2 8 42

10)2 3 0 1 0 1 5 1 0 0 1 2 3 3 6 3 28

10)3 2 3 1 3 3 1 0 0 0 0 8 3 9 6 3 42

10)4 0 3 1 0 1 2 0 1 0 1 6 11 5 1 2 34

10)5 0 0 0 1 0 1 0 0 0 2 3 9 1 0 1 18

10)6 1 0 0 0 0 1 0 0 0 0 0 1 1 0 0 4

Total no. of

positive samples

9 6 3 5 7 12 2 2 0 6 27 30 27 16 17 169

Total no.

of samples

20 21 19 17 18 17 19 18 19 20 33 33 32 29 25 340

Campylobacter spp. in cattle M. Hakkinen and M.-L. Hanninen

900 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905

ª 2009 The Authors

from which subtype S64 was isolated after two negative

samples.

Campylobacter jejuni was isolated from the drinking

trough at farm 3. The isolates represented the most

commonly detected SmaI subtypes (S17 and S70) in

that herd. No campylobacters were detected in the swab

samples from drinking troughs at farms 1 and 2.

No campylobacters were isolated from the milk samples.

Arcobacter butzleri, identified on the basis of aerobic growth

at 25�C, 37�C and 41�C, susceptibility to nalidixic acid and

resistance to cephalothin as well as the ability to hydrolyse

indoxyl acetate, was detected at a low level in three samples

from farm 3 (November 2006, April and May 2007) and in

one milk sample from farm 1 (April 2007).

Discussion

In our study, the prevalence of C. jejuni varied widely

between herds and sampling times. Atabay and Corry

(1998), Hanninen et al. (1998), Wesley et al. (2000) and

Nielsen (2002) have reported prevalences of C. jejuni

from 7% to 38% in dairy herds, which are similar to the

results from small herds 1 and 2 in our study. In herd 3

(the largest), the prevalence was high throughout the

study and was similar to prevalences reported among beef

cattle (Stanley et al. 1998; Besser et al. 2005). According

to Wesley et al. (2000), large herd size may contribute to

the transmission of infection because of the high number

of susceptible hosts that may be continuously challenged

by contact with carriers.

In herds 1 and 2, the prevalence of C. jejuni decreased

during the winter season, when the cattle were housed

indoors, which is similar to the results reported by

Hanninen et al. (1998). In contrast to their observation

of an increase in prevalence during the grazing period,

the overall prevalence in our study was lowest in August,

and herd 2 was campylobacter-negative, although the ani-

mals had been grazing since May. Lake water was likely

the origin of campylobacters in the study of Hanninen

et al. (1998), and Humphrey and Beckett (1987) also sug-

gested that natural waters were the main source of

campylobacters in cattle. The herds in our study had no

access to natural waters, which may explain the low prev-

alence in summer. Drinking water was obtained from the

well in the farm (herds 1 and 2) or from a municipal

source (herd 3). Despite the wide distribution of campy-

lobacters among wild birds and other animals and the

consequent contamination of the environment, the organ-

ism may not survive long enough to infect the grazing

cattle, except in water, where prolonged survival of

campylobacters has been reported (Cools et al. 2003).

However, after the grazing period in November 2007, the

Species

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

C. jejuni

Strain 100

806040

E1 2793/5/06 S7

S4

S75

S96

S97

S70

S94

S17

S95

S98

S101

S64

S100

E1 2938/1/07

E1 2016/1/06

E1 467/1/07

E1 2987/2/06

E1 1094/6/07

E1 3098/1/07

E1 3102/1/07

E1 2272/1/07

E1 4460/2/07

E1 4457/2/07

E1 2800/5/06

E1 2820/2/06

Type

Figure 2 Different SmaI PFGE profiles identified among Campylobacter jejuni isolates from three dairy cattle herds during the study period from

November 2006 to November 2007.

Table 2 Occurrence of SmaI PFGE subtypes of Campylobacter jejuni

in three dairy cattle herds on different sampling occasions between

November 2006 and November 2007

Sampling Herd 1 Herd 2 Herd 3

Nov 2006 S7, S64, S75 S4, S7 S17, S70

Feb 2007 S7, S64, S75 S4 S17, S70, S94

Apr 2007 S7, S64 S4 S17, S70, S95

Aug 2007 S7, S64, S96, S97 – S17, S70, S98, S101

Nov 2007 S7, S64, S96, S100 S7 S17, S70

M. Hakkinen and M.-L. Hanninen Campylobacter spp. in cattle

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905 901

prevalence was higher in all three herds than it was in

August. This difference could reflect a change in the diet,

as Stanley et al. (1998) suggested regarding seasonal peaks

in the excretion of campylobacters coinciding with the

beginning of the grazing period in spring and the transi-

tion to winter housing in autumn.

Contamination of milk can easily result from a lapse in

hygiene, when a herd is campylobacter-positive (Hum-

phrey and Beckett 1987; Schildt et al. 2005). The absence

of campylobacters in the milk samples indicates that all

the farms in the present study employed good milking

hygiene and were able to prevent faecal contamination of

the milk. Our detection of low-level contamination by

A. butzleri on four occasions is unexceptional, as other

studies focused specifically on Arcobacter have detected

the organism in raw tank milk as well (Scullion et al.

2006). Arcobacter butzleri is an environmental contami-

nant not directly associated with faecal contamination. It

has occasionally been isolated from human patients with

diarrhoea, but its significance as a food-borne pathogen is

under examination (Ho et al. 2006).

The prevalence of C. hyointestinalis ssp. hyointestinalis,

a common inhabitant of cattle (Atabay and Corry 1998;

Inglis et al. 2004; Hakkinen et al. 2007), is likely to be

higher in the herds than we detected. As the examination

focused on C. jejuni in this study, the detection method

for thermophilic campylobacters was chosen. Atabay and

Corry (1998) emphasized the significance of the choice of

medium when different Campylobacter species are the

organisms targeted for detection. Campylobacter coli was

infrequently detected in our samples. It is a minor species

in bovine intestines according to previous studies (Wesley

et al. 2000; Inglis et al. 2004; Hakkinen et al. 2007). In

most of the samples, only one Campylobacter species was

detected at a time, as Atabay and Corry (1998) also

reported. Catalase-negative Campylobacter sp. isolates

from farm 1 were probably C. sputorum biovar sputorum

(Vandamme and On 2001). Atabay and Corry (1998) iso-

lated similar unidentified catalase- and urease-negative,

H2S-producing campylobacters from cattle faeces.

We determined C. jejuni levels in the faecal samples by

using a semiquantitative method. Consequently, our

results are not fully comparable to counts from other

1 1

1

3 4

2

1

2

2 3

1 1

2 2

1 1

2

0

20

40

60

80

100 (a)

(b)

(c)

Sampling time

% o

f C. j

ejun

i iso

late

s %

of C

. jej

uni i

sola

tes

% o

f C. j

ejun

i iso

late

s

2

1

10

6

1

2

0

20

40

60

80

100

8

13 23 27 24

9

1 1 1 4 3 3

1 1

0

20

40

60

80

100

November 06 November 07February 07 April 07 August 07

Sampling time November 06 November 07 February 07 April 07 August 07

Sampling time November 06 November 07February 07 April 07 August 07

Figure 3 Percentages of Campylobacter jejuni SmaI subtypes in three

dairy cattle herds on different sampling occasions between November

2006 and November 2007. Herd 1 (a): S7; S64; S75; S96;

S97; S100, Herd 2 (b): S4 ; S7, and Herd 3 (c) S17;

S70; S94; S95; S98; S101. The number of isolates repre-

senting each subtype is indicated on top of each column.

Table 3 Occurrence of Campylobacter jejuni SmaI subtypes in the

individual animals of Herd 1 on different sampling occasions between

November 2006 and November 2007

Animal

no.

Sampling time

Nov 2006 Feb 2007 Apr 2007 Aug 2007 Nov 2007

55 Neg* Neg Neg Neg Neg

107 Neg Neg Neg S100 S100

108 Neg Neg Neg Neg S96

109 Neg Neg Neg Neg ND�

124 S64 S64 S64 Neg Neg

128 Neg Neg Neg ND S7

129 Neg Neg Neg Neg ND

130 ND Neg Neg Neg Neg

131 S64 S64 S64 S64 S64

134 S7 S7 S7 S7 Neg

136 S7 Neg Neg Neg Neg

139 Neg Neg Neg S97 Neg

140 S7 S7 Neg Neg S64

141 Neg Neg Neg Neg Neg

*Not detected.

�Not done.

Campylobacter spp. in cattle M. Hakkinen and M.-L. Hanninen

902 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905

ª 2009 The Authors

studies. However, C. jejuni levels in the positive samples

were generally low. In 13% of the samples, C. jejuni was

detected in dilution 10)5 or higher, indicating levels that

are closer to counts reported from beef cattle and calves

(Stanley et al. 1998; Nielsen 2002; Inglis et al. 2004) than

the average concentrations of 1Æ2 · 102 CFU g)1 and 69

MPN g)1 in dairy cattle reported by Nielsen (2002) and

Stanley et al. (1998), respectively. The C. jejuni levels in

our study varied depending on the herd and sampling

time. Stanley et al. (1998) observed a clear seasonality,

with spring and autumn peaks in the number of thermo-

philic campylobacters in the faeces of dairy herds. As a

result of the small number of herds and only a 1-year

sampling period, such conclusions cannot be drawn from

our study, where opposite trends in prevalences as well as

in C. jejuni levels in positive animals occurred among the

herds.

The detection of only few PFGE subtypes in each herd

may reflect a small number of sources of C. jejuni and

transmission of the organism between animals in the herd

rather than the introduction of new types from various

sources. Of the seven and six different SmaI subtypes of

C. jejuni identified in herds 1 and 3, respectively,

two subtypes persisted in each of the herds throughout

the 1-year sampling period. In August and November

2007, new subtypes also emerged, possibly from the

environment during grazing. Until the end of the study,

however, these new subtypes were unable to exclude the

original ones and seemed to be only temporarily excreted,

as three of the four new types found in August were no

longer detectable in November. The persistent subtypes

may represent genotypes especially adapted to colonizing

bovine intestines. Two of them, S7 and S64, were identi-

cal with the SmaI types commonly detected in human

campylobacter infections of domestic origin as well as in

chicken flocks in Finland (M. Hakkinen et al. unpub-

lished data) and seem able to colonize diverse hosts. The

other two persistent subtypes may represent host-specific

C. jejuni genotypes in cattle, which may not pose a signi-

ficant health risk to humans (Karenlampi et al. 2007;

McCarthy et al. 2007). In herd 2, only two subtypes (S4

and S7) were detected during the entire sampling period.

Subtype S7 was found only on two occasions: the first

and last samplings. This may, however, indicate that this

subtype existed in the herd permanently, perhaps at con-

centrations below the detection limit or in the intestines

of individual animals from which samples were not

obtained on all occasions.

Only one C. jejuni subtype was detected in most of the

samples, when two to six isolates per sample were analy-

sed with PFGE. This is consistent with sero- and genotyp-

ing results from other studies (Hanninen et al. 1998;

Nielsen 2002). The results of individual animals suggest

that the shedding of campylobacters is principally inter-

mittent, but the amount of campylobacters excreted can

be occasionally high. One of the animals in herd 1 could

be considered as a permanent carrier, as the same subtype

of C. jejuni was isolated on all sampling occasions from

its faeces. Hanninen et al. (1998) have reported similar

observations on the persistence of C. jejuni in the intes-

tines of an individual animal. On the contrary, some of

the animals in our study were campylobacter-negative on

all sampling occasions, suggesting that permanently

campylobacter-negative animals may also exist, although

a higher frequency of sampling and an extended sampling

period could also have yielded some positive samples

from these individuals. However, animal-related factors

may exist that render some individuals more resistant

than others to campylobacter colonization.

No samples of drinking water were examined in this

study. However, persons living on the farms consumed

water drawn from the same sources with no intestinal

disturbances. The detection of the same subtypes of

C. jejuni from the faecal samples and the surface sam-

ple from the drinking trough of herd 3 suggests that

the trough was contaminated by faeces and could circu-

late campylobacters among the animals of the herd.

Water trough surface contaminated by campylobacters

was implicated as a risk factor in the study by Minihan

et al. (2004).

The results of this study suggest that permanent or

long-term shedding of the same subtypes of C. jejuni

occurs in dairy cattle. In addition, after initial coloniza-

tion of the gut by one subtype, no other subtype may be

able to exclude it at a later stage. Kwan et al. (2008) have

also presented results suggesting ecological competition

between campylobacter strains in bovine gut. The

decreasing prevalence of campylobacters in herds without

access to natural water sources during summer grazing

can be considered indirect evidence of the significance of

drinking water in the transmission of campylobacters.

The ability of different C. jejuni subtypes to colonize

bovine intestines may vary, and individual resistance to

Campylobacter colonization may differ between animals in

a cattle herd as well. Moreover, at least some subtypes

common in human infections may be permanent or long-

term colonizers in the bovine gut.

Acknowledgements

This study was partially funded by the Ministry of Agri-

culture and Forestry and Walter Ehrstrom Foundation.

The authors would like to thank Kirsi-Maria Eklund, Sari

Maljanen and Mirva Tahvanainen for their technical assis-

tance. They would also like to thank the farmers for their

cooperation.

M. Hakkinen and M.-L. Hanninen Campylobacter spp. in cattle

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905 903

References

Anonymous (2006) International Standard ISO 10272-1. Micro-

biology of Food and Animal Feeding Stuffs – Horizontal

Method for Detection and Enumeration of Campylobacter

spp. Part 1: Detection Method. Geneva, Switzerland: Inter-

national Organization for Standardization (ISO).

Anonymous (2007a) The Community summary report on trends

and sources of zoonoses, zoonotic agents, antimicrobial resis-

tance and foodborne outbreaks in the European Union in

2006. EFSA J 130, 106–352.

Anonymous (2007b) Thermotolerant Campylobacter. Detection,

Semi-quantitative and Quantitative Determination in Foods

and Drinking Water, Method No. 119, 3. Oslo, Norway:

Nordic Committee on Food Analysis.

Atabay, H.I. and Corry, J.E.L. (1998) The isolation and preva-

lence of campylobacters from dairy cattle using a variety

of methods. J Appl Microbiol 84, 733–740.

Besser, T.E., LeJeune, J.T., Rice, D.H., Berg, J., Stilborn, R.P.,

Kaya, K., Bae, W. and Hancock, D.D. (2005) Increasing

prevalence of Campylobacter jejuni in feedlot cattle through

the feeding period. Appl Environ Microbiol 71, 5752–5758.

Clark, G.C., Price, L., Ahmed, R., Woodward, D., Melito, P.L.,

Rodgers, F.G., Jamieson, F., Ciebin, B. et al. (2003) Char-

acterization of water-borne outbreak-associated Campylo-

bacter jejuni, Walkerton, Ontario. Emerg Infect Dis 9,

1232–1241.

Cools, I., Uyttendale, M., Caro, C., D’Haese, E., Nelis, H.J.

and Debevere, J. (2003) Survival of Campylobacter jejuni

strains of different origin in drinking water. J Appl Micro-

biol 94, 886–892.

Devane, M.L., Nicol, C., Ball, A., Klena, J.D., Scholes, P.,

Hudson, J.A., Baker, M.G., Gilpin, B.J. et al. (2005) The

occurrence of Campylobacter subtypes in environmental

reservoirs and potential transmission routes. J Appl Micro-

biol 98, 980–990.

Evans, M.R., Roberts, R.J., Ribeiro, C.D., Gardner, D. and

Kembrey, D. (1996) A milk-borne campylobacter outbreak

following an educational farm visit. Epidemiol Infect 117,

457–462.

Garrett, N., Devane, M.L., Hudson, J.A., Nicol, C., Ball, A.,

Klena, J.D., Scholes, P., Baker, M.G. et al. (2007) Statistical

comparison of Campylobacter jejuni subtypes from human

cases and environmental sources. J Appl Microbiol 103,

2113–2121.

Gormley, F.J., MacRae, M., Forbes, K.J., Ogden, I.D., Dallas,

J.F. and Strachan, N.J.C. (2008) Has retail chicken played

a role in the decline of human campylobacteriosis. Appl

Environ Microbiol 74, 383–390.

Hakkinen, M., Heiska, H. and Hanninen, M.-L. (2007) Preva-

lence of Campylobacter spp. in cattle in Finland and anti-

microbial susceptibility of bovine Campylobacter jejuni

strains. Appl Environ Microbiol 73, 3232–3238.

Hanninen, M.-L., Niskanen, M. and Korhonen, L. (1998)

Water as a reservoir for Campylobacter jejuni infection in

cows studied by serotyping and pulsed-field gel electropho-

resis. J Vet Med B 45, 37–42.

Ho, H.T., Lipman, L.J. and Gaastra, W. (2006) Arcobacter,

what is known and unknown about a potential foodborne

zoonotic agent! Vet Microbiol 115, 1–13.

Humphrey, T.J. and Beckett, B. (1987) Campylobacter jejuni

in dairy cows and raw milk. Epidemiol Infect 98, 263–

269.

Inglis, G.D., Kalischuk, L.D. and Busz, H.W. (2004) Chronic

shedding of Campylobacter species in beef cattle. J Appl

Microbiol 97, 410–420.

Karenlampi, R., Rautelin, H., Schonberg-Norio, D., Paulin,

L. and Hanninen, M.-L. (2007) Longitudinal study of

Finnish Campylobacter jejuni and C. coli isolates from

humans, using multilocus sequence typing, including

comparison with epidemiological data and isolates

from poultry and cattle. Appl Environ Microbiol 73,

148–155.

Kwan, P.L.S., Birtles, A., Bolton, F.J., French, N.P., Robinson,

S.E., Newbold, L.S., Upton, M. and Fox, A.J. (2008) Longi-

tudinal study of the molecular epidemiology of Campylo-

bacter jejuni in cattle on dairy farms. Appl Environ

Microbiol 74, 3626–3633.

Lehner, A., Schneck, C., Feierl, G., Pless, P., Deutz, A., Brandl,

E. and Wagner, M. (2000) Epidemiologic application of

pulsed-field gel electrophoresis to an outbreak of Campylo-

bacter jejuni in an Austrian youth centre. Epidemiol Infect

125, 13–16.

McCarthy, N., Colles, F.M., Dingle, K.E., Bagnall, M.C.,

Manning, G., Maiden, M.C.J. and Falush, D. (2007) Host-

associated genetic import in Campylobacter jejuni. Emerg

Infect Dis 13, 267–272.

Minihan, D., Whyte, P., O’Mahony, M., Fanning, S., McGill,

K. and Collins, J.D. (2004) Campylobacter spp. in Irish

feedlot cattle: a longitudinal study involving pre-harvest

and harvest phases of the food chain. J Vet Med B 51,

28–33.

Neimann, J., Engberg, J., Mølbak, K. and Wegener, H.C.

(2003) A case–control study of risk factors for sporadic

campylobacter infections in Denmark. Epidemiol Infect

130, 353–366.

Nielsen, E.M. (2002) Occurrence and strain diversity of ther-

mophilic campylobacters in cattle of different age groups

in dairy herds. Lett Appl Microbiol 35, 85–89.

Schildt, M., Savolainen, S. and Hanninen, M.-L. (2005)

Long-lasting Campylobacter jejuni contamination of milk

associated with gastrointestinal illness in a farming family.

Epidemiol Infect 134, 401–405.

Scullion, R., Harrington, C.S. and Madden, R.H. (2006) Preva-

lence of Arcobacter spp. in raw milk and retail raw meats

in Northern Ireland. J Food Prot 69, 1986–1090.

Stanley, K.N., Wallace, J.S., Currie, J.E., Diggle, P.J. and Jones,

K. (1998) The seasonal variation of thermophilic campy-

lobacters in beef cattle, dairy cattle and calves. J Appl

Microbiol 85, 472–480.

Campylobacter spp. in cattle M. Hakkinen and M.-L. Hanninen

904 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905

ª 2009 The Authors

Studahl, A. and Andersson, Y. (2000) Risk factors for indige-

nous campylobacter infection: a Swedish case control

study. Epidemiol Infect 125, 269–275.

Vandamme, P. and On, S.L.W. (2001) Recommendations of

the subcommittee on the taxonomy of Campylobacter

and related bacteria. Int J Syst Evol Microbiol 51, 719–

721.

Wesley, I.V., Wells, S.J., Harmon, K.M., Green, A.,

Schroeder-Tucker, L., Glover, M. and Siddique, I. (2000)

Fecal shedding of Campylobacter and Arcobacter

spp. in dairy cattle. Appl Environ Microbiol 66, 1994–

2000.

Whyte, P., McGill, K., Cowley, D., Madden, R.H., Moran, L.,

Scates, P., Carroll, C., O’Leary, A. et al. (2004) Occurrence

of Campylobacter in retail foods in Ireland. Int J Food

Microbiol 95, 111–118.

Wingstrand, A., Neimann, J., Engberg, J., Nielsen, E.M.,

Gerner-Schmidt, P., Wegener, H. and Mølbak, K. (2006)

Fresh chicken as main risk factor for campylobacteriosis,

Denmark. Emerg Infect Dis 12, 280–284.

M. Hakkinen and M.-L. Hanninen Campylobacter spp. in cattle

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 898–905 905

III

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Feb. 2009, p. 1208–1210 Vol. 75, No. 40099-2240/09/$08.00�0 doi:10.1128/AEM.01879-08Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Bovine Campylobacter jejuni Strains Differ from Human and ChickenStrains in an Analysis of Certain Molecular Genetic Markers�

Manuel Gonzalez,1 Marjaana Hakkinen,2 Hilpi Rautelin,3,4 and Marja-Liisa Hanninen1*Department of Food and Environmental Hygiene, University of Helsinki,1 Microbiology Unit, Research Department,

Finnish Food Safety Authority,2 Department of Bacteriology and Immunology, Haartman Institute, University ofHelsinki,3 and HUSLAB, Helsinki University Central Hospital Laboratory,4 Helsinki, Finland

Received 13 August 2008/Accepted 14 December 2008

The association of four new genetic markers with a chicken, bovine, or human host was studied among 645Campylobacter jejuni isolates. The �-glutamate transpeptidase gene and dmsA were common in human andchicken isolates but uncommon among bovine isolates. In the t test, bovine isolates differed significantly (P <0.05) from human and chicken isolates.

Campylobacter jejuni is a zoonotic human enteric pathogenwith a large number of animal hosts (12, 19). Campylobacte-riosis is a leading cause of human bacterial gastroenteritis inmany industrialized countries (19). Epidemiological studies in-dicate that exposure to improperly cooked chicken meat, han-dling of raw chicken meat, and drinking unpasteurized milk areimportant risk factors for campylobacteriosis (12, 15, 19, 20).

The role of different animal sources in human infections isnot well characterized. Molecular typing methods applied forfingerprinting of C. jejuni strains have shown overlapping ge-notypes between animal and human isolates (5, 16, 17, 21).Population biological studies using multilocus sequence typing(6) have revealed that a host-C. jejuni interaction may leave asignature in the bacterial genome. As a consequence, e.g.,chicken- or cattle-associated populations can be assigned totheir hosts (18). We investigated host association of C. jejuniisolates from cattle, chickens, and humans using PCR detec-tion of four new genetic markers developed under our study.Using comparative genomics (3), four genetic markers—i.e.,ggt, the �-glutamyl transpeptidase gene; dmsA (Cju34), a sub-unit of the putative tripartite anaerobic dimethyl sulfoxide(DMSO) oxidoreductase (DMSO/trimethylamine N-oxide re-ductase) gene; Cj1585c, coding for a putative oxidoreductase;and CJJ81176-1371, a putative serine protease gene—wereselected from the genomes of C. jejuni strains 81-176 (10),RM1221, and NCTC 11168. ggt is in the genome of 81-176 butnot in the genome of NCTC 11168 or RM1221 (10). GeneCj1585c of NCTC 11168 is replaced in 81-176 by a cluster offour genes (dmsA, dmsB, dmsC, and dmsD) (10). The presenceof these four genes in a total of 645 C. jejuni isolates frombovine fecal samples (n � 131) (8), chicken cecal or meatsamples (n � 205), and human patients (n � 309) (16, 17) wasexamined by PCR to find their suitability for host associationstudies. PCR primers designed for the amplification of thefragments are shown in Table 1. Twelve PCR products for eachgene fragment were sequenced. The sequences of each gene

were shown to be rather conserved (95.5 to 100% similaritywithin each gene) because only a few nucleotide positions(from 2 to 9) were found to be variable.

Statistical analyses were performed using SPSS software.The �2 test was used to test for similarity in the frequencies ofmarker genes within the isolates from different hosts. In addi-tion, we used the paired two-tailed Student’s t test for analysisof host associations for the combined set of four genes.

Frequencies of the genes are shown in Table 2. Similarly, theresults of the paired two-tailed t test on the significance of thefrequencies of the combined four genes from different hostsare shown in Table 2. These results indicated significant (P �0.05) association of bovine and chicken isolates with their hostsource, but a high similarity was observed between the chickenand human isolates (P � 0.9949). Annual frequencies of thegenes are presented for human isolates in Table 3 and forchicken isolates in Table 4. The analysis of the annual frequen-cies of the four genes combined showed that the human iso-lates were similar in 1996 and 2002 and 2002 and 2003, butdiffered between 1996 and 2003 (Table 3). The chicken isolateswere similar in all study years (Table 4). These results revealedthat these genes associated with metabolism and energy pro-duction (ggt, oxidoreductases) (2, 11, 22), colonization (ggt) (2,11), or unknown function (serine protease genes) are not ran-domly distributed among the isolates from different hosts butshow a host association.

* Corresponding author. Mailing address: Department of Food andEnvironmental Hygiene, P.O. Box 66, 0014 University of Helsinki,Finland. Phone: 358-9-19157113. Fax: 358-9-19157101. E-mail: marja-liisa.hanninen@helsinki.fi.

� Published ahead of print on 19 December 2008.

TABLE 1. PCR primers used amplification of the fragments of thefour marker genes

Gene (product)Primer sequence Product

size(bp)Forward Reverse

ggt (�-glutamyltranspeptidase)

TTTTAGCCATATCCGCTGCT

AGCTGGAGTACCAGGAA

339

dmsAa GATAGGGCATTGCGATGAGT

CTTGCTAGCCCAATCAGGAG

238

Cj1585c(oxidoreductase)

TGTTGTGGGTTTGCTGGATA

TTGCTTCACTGCATTCATCC

202

CJJ81176-1367/1371(serine protease)

TGCAAAGCAGGGCTAAGAAT

TTATGGAGCTGGGGTGTTTC

318

a Cju34.

1208

The intestinal environments of cattle and chicken are quitedifferent, which may select isolates with variable characteris-tics, e.g., related to energy metabolism, adaptation to lower orhigher oxygen contents or amino acid metabolism. C. jejunicolonization in dairy cattle can be persistent, as shown by thestudies in which the same genotype was isolated for up to 1year (1, 13, 14). The life cycle of cattle is several years, pro-viding a long potential time span for the adaptation of C. jejuniwith its host. The life cycle of chickens, in contrast, is muchshorter, 5 weeks or more. Our results suggested that hostadaptation of certain C. jejuni strains is evident. The dmsAsubunit was more often detected among chicken and humanisolates than among bovine isolates (Table 2). In addition,dmsA-positive chicken isolates occurred with similar high an-nual frequency in 2003, 2006, and 2007 (Table 4), indicatingthat this characteristic is most probably important in coloniza-tion. The occasional significant annual fluctuation seen in thefrequency of dmsA-positive human isolates may reflect varia-tion in the infection sources (Table 3). In a recent study (9),dmsB was one of the genes present in C. jejuni strain A 74/C,shown to be robust colonizer in chickens, but absent from C.jejuni 11168(GS), a poorly colonizing strain (7). The C. jejuniNCTC 11168, 81116, and 81-176 strains have another putativeDMSO oxidoreductase gene (homologous to Cj0264c) thatdiffers from Cju34. In opposition, the Cj1585c-type oxi-doreductase was more frequently present in isolates from cat-tle than in those from chickens or humans (Table 2). Analysesof C. jejuni genomes have predicted a branched complex elec-tron transport chain capable of utilizing multiple electron do-

nors and acceptors (22), and our results suggest flexibility inthe oxidoreductase systems as well.

ggt (�-glutamyl transpeptidase) has been shown to be impor-tant in the persistent colonization of C. jejuni in chickens (2),and recent studies (11) further extend the significance of thisgene in the glutamine and glutathione metabolism and colo-nization of C. jejuni. In our study, the frequency of the ggt-positive human and chicken isolates was high (Table 2) and thefrequencies remained similar over the study years (Tables 3and 4). These results further reveal the importance of �-glu-tamyl transpeptidase in colonization and pathogenesis. In con-trast, a low frequency of ggt-positive isolates (8.4%) was foundamong bovine isolates (Table 2), suggesting that this type ofmetabolism is not crucial for colonization of the bovine gut.Similar variable frequencies to those in our study were foundin the study by Barnes et al. (2).

The genomes of NCTC 11168, RM1221, and 81-176 have asubtilase-type serine protease gene homologous to CJJ81176-1367, which is located close to the CJJ81176-1371 gene in thegenome of 81-176 (10). The G�C composition of this gene is29%, whereas the G�C composition of CJJ81176-1371 is 36%,indicating that these genes most probably have different evo-lutionary origins. In our study, the serine gene was commonamong bovine isolates (Table 2) and less common amongchicken and human isolates. The primers we used may amplifyboth types of the subtilase genes. Proteases in C. jejuni have arole in stress tolerance (4). Whether the serine protease isimportant in the pathogenesis of campylobacteriosis remainsto be elucidated.

TABLE 3. Frequency of the four marker genes ggt, Cj1585c, dmsA (Cju34), and CJJ81176-1367/1371 in 309 C. jejuni isolates from humans

Marker gene (product)No. of isolates with gene/total no. of isolates (%) P value for yr:

1996 2002 2003 1996–2002 1996–2003 2002–2003

�2 testggt (�-glutamyl transpeptidase) 52/97 (53.6) 57/111 (51.3) 60/101 (59.4) 0.74 0.41 0.24Cj1585c (oxidoreductase) 27/97 (27.8) 25/111 (22.5) 47/101 (46.5) 0.38 �0.05 �0.05dmsAa 69/97 (71.3) 101/111 (91) 86/101 (85.1) �0.05 �0.05 0.19CJJ81176-1367/1371 (serine protease) 34/97 (35.1) 37/111 (33.3) 46/101 (45.5) 0.79 0.13 0.07

t testb

0.4506 0.0003 0.052

a dmsA (Cju34) is a subunit of the putative tripartite anaerobic DMSO oxidoreductase gene.b Significance (P � 0.05) of the frequency of the combined four genes.

TABLE 2. Frequency of the four marker genes ggt, Cj1585c, dmsA (Cju34), and CJJ81176-1371 in 645 human, chicken, and cattle C.jejuni isolates

Marker gene (product)No. of isolates with gene/total no. of isolates (%) P value for sourcea:

Human Chicken Bovine Human/chicken Chicken/bovine Human/bovine

�2 testggt (�-glutamyl transpeptidase) 169/309 (54.7) 75/205 (36.6) 11/131 (8.4) �0.05 �0.05 �0.05Cj1585c (oxidoreductase) 99/309 (32) 49/205 (23.9) 83/131 (62.6) �0.05 �0.05 �0.05dmsAb 256/309 (82.8) 151/205 (73.3) 18/131 (13.7) �0.05 �0.05 �0.05CJJ81176-1367/1371 (serine protease) 117/309 (37.8) 74/205 (36.1) 96/131 (73.3) 0.68 �0.05 �0.05

t testc

0.9949 0.0087 0.0122

a P � 0.05 represents significant difference.b dmsA (Cju34) is a subunit of the putative tripartite anaerobic DMSO oxidoreductase gene.c Significance (P � 0.05) of the frequency of the combined four genes by paired two-tailed t test.

VOL. 75, 2009 CAMPYLOBACTER JEJUNI MARKER GENE-HOST ASSOCIATION 1209

The genetic markers associated with metabolism, coloniza-tion, or an unknown protease function allowed assignment ofthe chicken or bovine source of C. jejuni. These results suggestthat metabolic diversity is an important adaptive factor in hostadaptation.

We acknowledge financial support from the Academy of Finland(Elvira) and EU project no. 036272 (Biotracer).

REFERENCES

1. Bae, W., K. N. Kaya, D. D. Hancock, D. R. Call, Y. H. Park, and T. E. Besser.2005. Prevalence and antimicrobial resistance of thermophilic Campylobacterspp. from cattle farms in Washington State. Appl. Environ. Microbiol. 71:169–174.

2. Barnes, I. H., M. C. Bagnall, D. D. Browning, S. A. Thompson, G. Manning,and D. G. Newell. 2007. Gamma-glutamyl transpeptidase has a role in thepersistent colonization of the avian gut by Campylobacter jejuni. Microb.Pathog. 43:198–207.

3. Chaudhuri, P. C., and M. J. Pallen. 2006. xBASE, a collection of onlinedatabases for bacterial comparative genomics. Nucleic Acids Res. 34:D335–D337.

4. Cohn, M. T., H. Ingmer, F. Mulholland, K. Jørgensen, J. M. Wells, and L.Brøndsted. 2007. Contribution of conserved ATP-dependent proteases ofCampylobacter jejuni to stress tolerance and virulence. Appl. Environ. Mi-crobiol. 73:7803–7813.

5. Denis, M., B. Chidaine, M. J. Laisney, I. Kempf, K. Rivoal, F. Megraud, andP. Fravalo. 2008. Comparison of genetic profiles of Campylobacter strainsisolated from poultry, pig and Campylobacter human infections in Brittany,France. Pathol. Biol. (Paris). doi:10.1016/j.patbio.2008.04007.

6. Dingle, K. E., F. M. Colles, D. R. A. Wareing, R. Ure, A. J. Fox, F. E. Bolton,H. J. Bootsma, R. J. L. Willems, R. Urwin, and M. C. J. Maiden. 2001.Multilocus sequence typing system for Campylobacter jejuni. J. Clin. Micro-biol. 39:14–23.

7. Gaynor, E. C., S. Cawthraw, G. Manning, J. K. MacKichan, S. Falkow, andD. G. Newell. 2004. The genome-sequenced variant of Campylobacter jejuniNCTC 11168 and the original clonal clinical isolate differ markedly in colo-nization, gene expression, and virulence-associated phenotypes. J. Bacteriol.186:503–517.

8. Hakkinen, M., H. Heiska, and M.-L. Hanninen. 2007. Prevalence of Campy-lobacter spp. in cattle in Finland and antimicrobial susceptibility of bovineCampylobacter jejuni strains. Appl. Environ. Microbiol. 73:3232–3238.

9. Hiett, K. L., A. Stintzi, T. M. Andacht, R. L. Kuntz, and B. S. Seal. 2008.Genomic differences between Campylobacter jejuni isolates identify surfacemembrane and flagellar function gene products potentially important forcolonizing the chicken intestine. Funct. Integr. Genomics 8:407–420.

10. Hofreuter, D., J. Tsai, R. O. Watson, V. Novik, B. Altman, M. Benitez, C.Clark, C. Perbost, T. Jarvie, L. Du, and J. E. Galan. 2006. Unique featuresof a highly pathogenic Campylobacter jejuni strain. Infect. Immun. 74:4694–4707.

11. Hofreuter, D., V. Novik, and J. E. Galan. 2008. Metabolic diversity in Campy-lobacter jejuni enhances specific tissue colonization. Cell Host Microbe4:425–433.

12. Humphrey, T., S. O’Brien, and M. Madsen. 2007. Campylobacters as zoo-notic pathogens: a food production perspective. Int. J. Food Microbiol.117:237–257.

13. Inglis, G. D., L. D. Kalischuk, and H. W. Busz. 2004. Chronic shedding ofCampylobacter species in beef cattle. J. Appl. Microbiol. 97:410–420.

14. Johnsen, G., K. Zimmerman, B. A. Lindstedt, T. Vardund, H. Herikstad, andG. Kapperud. 2006. Intestinal carriage of Campylobacter jejuni and Campy-lobacter coli among cattle from south-western Norway and comparativegenotyping of bovine and human isolates by amplified-fragment length poly-morphism. Acta Vet. Scand. 48:4–9.

15. Kapperud, G., G. Espeland, E. Wahl, A. Walde, H. Herikstad, S. Gustavsen,I. Tveit, O. Natås, L. Bevanger, and A. Digranes. 2003. Factors associatedwith increased and decreased risk of Campylobacter infection: a prospectivecase-control study in Norway. Am. J. Epidemiol. 158:234–242.

16. Karenlampi, R., H. Rautelin, M. Hakkinen, and M.-L. Hanninen. 2003.Temporal and geographical distribution and overlap of Penner heat-stableserotypes and pulsed-field electrophoresis genotypes of Campylobacter jejuniisolates collected from humans and chickens in Finland during a seasonalpeak. J. Clin. Microbiol. 41:4870–4872.

17. Karenlampi, R., H. Rautelin, D. Schonberg-Norio, L. Paulin, and M.-L.Hanninen. 2007. Longitudinal study of Finnish Campylbacter jejuni andCampylobacter coli isolates from humans, using multilocus sequence typing,including comparison with epidemiological data and isolates from poultryand cattle. Appl. Environ. Microbiol. 73:148–155.

18. McCarthy, N. D., F. M. Colles, K. E. Dingle, M. C. Bagnall, G. Manning,M. C. Maiden, and D. Falush. 2007. Host-associated genetic import inCampylobacter jejuni. Emerg. Infect. Dis. 13:267–272.

19. Olson, K. C., S. Ethelberg, W. van Pelt, and R. V. Tauxe. 2008. Epidemiologyof Campylobacter jejuni infections in industrialized nations, p. 163–189. In I.Nachamkin, C. M. Szymanski, and M. J. Blaser (ed.), Campylobacter, 3rd ed.American Society for Microbiology, Washington, DC.

20. Schonberg-Norio, D., J. Takkinen, M.-L. Hanninen, M.-L. Katila, S. S.Kaukoranta, L. Mattila, and H. Rautelin. 2004. Swimming and Campy-lobacter infections. Emerg. Infect. Dis. 10:1474–1477.

21. Wassenaar, T. M., and D. G. Newell. 2000. Genotyping of Campylobacterspp. Appl. Environ. Microbiol. 66:1–9.

22. Weingarten, R. A., J. L. Grimes, and J. W. Olson. 2008. Role of Campy-lobacter jejuni respiratory oxidases and reductases in host colonization. Appl.Environ. Microbiol. 74:1367–1375.

TABLE 4. Presence of the four marker genes ggt, Cj1585c, dmsA (Cju34), and CJJ81176-1367/1371 in 205 C. jejuni isolates from chickens

Marker gene (product)No. of isolates with gene/total no. of isolates (%) P value for yr:

2003 2006 2007 2003–2006 2003–2007 2006–2007

�2 testggt (�-glutamyl transpeptidase) 16/37 (43.2) 29/71 (40.8) 30/97 (30.9) 0.81 0.19 0.18Cj1585c (oxidoreductase) 15/37 (40.5) 6/71 (8.5) 28/97 (28.9) �0.05 0.21 �0.05dmsAa 30/37 (81.1) 49/71 (69) 72/97 (74.2) 0.15 0.38 0.46CJJ81176-1367/1371 (serine protease) 20/37 (54.1) 23/71 (32.4) 31/97 (31.9) �0.05 �0.05 0.95

t testb

0.074 0.095 0.317

a dmsA (Cju34) is a subunit of the putative tripartite anaerobic DMSO oxidoreductase gene.b Significance (P � 0.05) of the frequency of the combined four genes.

1210 GONZALEZ ET AL. APPL. ENVIRON. MICROBIOL.

IV

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 2009, p. 5244–5249 Vol. 75, No. 160099-2240/09/$08.00�0 doi:10.1128/AEM.00374-09Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Chickens and Cattle as Sources of Sporadic Domestically AcquiredCampylobacter jejuni Infections in Finland�

Marjaana Hakkinen,1,2* Ulla-Maija Nakari,3 and Anja Siitonen3

Research Department, Finnish Food Safety Authority Evira, Helsinki, Finland1; Department of Food and Environmental Hygiene,FI-00014 University of Helsinki, Helsinki, Finland2; and Unit of Gastrointestinal Infections, National Institute for Health and

Welfare, Helsinki, Finland3

Received 16 February 2009/Accepted 12 June 2009

A substantial sampling among domestic human campylobacter cases, chicken process lots, and cattle atslaughter was performed during the seasonal peak of human infections. Campylobacter jejuni isolates (n � 419)were subtyped using pulsed-field gel electrophoresis with SmaI, and isolates representing overlapping types(n � 212) were further subtyped using KpnI for restriction. The SmaI/KpnI profiles of 55.4% (97/175) of thehuman isolates were indistinguishable from those of the chicken or cattle isolates. The overlapping SmaI/KpnIsubtypes accounted for 69.8% (30/43) and 15.9% (32/201) of the chicken and cattle isolates, respectively. Theoccurrence of identical SmaI/KpnI subtypes with human C. jejuni isolates was significantly associated withanimal host species (P < 0.001). A temporal association of isolates from chickens and patients was possible in31.4% (55/175) of the human infections. Besides chickens as sources of C. jejuni in the sporadic infections, therole of cattle appears notable. New approaches to restrict the occurrence of campylobacters in other farmanimals may be needed in addition to hygienic measures in chicken production. However, only about half of thehuman infections were attributable to these sources.

The incidence of human enteric infections caused by campy-lobacters is highest in the summer months, showing a consis-tent peak at the end of July in Finland (www.ktl.fi/attachments/suomi/julkaisut/julkaisusarja_b/2008/2008b09.pdf), as well asin other Nordic countries (16, 33). Almost 70% of campy-lobacter infections detected in July and August in Finland aredomestically acquired, whereas the annual average proportionof domestic cases is about 30%, and most of them are causedby Campylobacter jejuni (30). The prevalence of campylobactersin Finnish broiler flocks peaks simultaneously with the humancases (7), and similar sero- and genotypes have been reportedamong human and poultry strains isolated in Finland and inother countries (5, 8, 21–23). Several epidemiological studieshave identified the handling and consumption of raw or un-dercooked poultry meat as a major risk factor for campy-lobacter enteritis (for example, see references 18, 20, and 41),whereas opposite conclusions about the significance of theconsumption of chicken meat were drawn from the Swedishcase-control study among young children (2) and an extensiveDanish register-based study (6).

Data derived from the genotyping studies of C. jejuni isolatesfrom human infections and animals support the currentsuggestion that poultry is the most important single sourceof sporadic campylobacteriosis (12, 22, 29). However, sev-eral reports on genotype comparisons suggest that poultrymay be a less significant source of campylobacters thangenerally thought, and other animal reservoirs should also beconsidered notable sources of campylobacters pathogenic tohumans (3, 8, 17, 27, 31). Studies of the temporal occurrence of

campylobacters in human infections and poultry flocks haverevealed that the peak in prevalence, as well as some of theoverlapping sero- and genotypes, is detected in humans priorto being detected in poultry (21, 28).

Although cattle are well-known carriers of campylobacters,the survival of these fragile organisms in beef is poor (39, 42).In recent years, some authors (1, 4, 10) have raised the ques-tion of an indirect association between cattle and human cases.In a Finnish study combining data from the multilocus sequencetyping of campylobacters isolated from production animals andfrom epidemiological studies of human cases, significant associa-tions emerged between certain sequence-type complexes fromhuman infections and contact with cattle, the consumption ofunpasteurized milk, or the tasting or consumption of raw mincedmeat (23).

The aim of this study was to investigate the contributions ofpoultry and cattle as sources of human C. jejuni infections inFinland by comparing over a limited time frame the macro-restriction profiles obtained from pulsed-field gel electro-phoresis (PFGE) analysis of a geographically representativecollection of C. jejuni isolates from domestically acquired spo-radic human infections, chicken process lots, and cattle.

MATERIALS AND METHODS

Isolates. We studied a total of 419 isolates. Human C. jejuni isolates (n � 175)were collected from June to August 2003, during the seasonal peak of humancases. The isolates represented all domestic C. jejuni strains isolated in 9 of 25clinical microbiology laboratories located in nine hospital districts across thecountry. They were isolated from the fecal samples of patients using modifiedcharcoal cefoperazone deoxycholate agar. One isolate per patient was submittedto the National Public Health Institute (KTL; currently, the National Institutefor Health and Welfare [THL]) for further investigation, and an isolate wasdefined as domestic if the patient had no history of traveling abroad within 10days before the onset of symptoms or 17 days before the specimen was taken.Only isolates from sporadic infections were included.

Bovine fecal (n � 186) and carcass (n � 15) isolates were obtained from

* Corresponding author. Mailing address: Finnish Food Safety Au-thority Evira, Mustialankatu 3, FI-00790 Helsinki, Finland. Phone: 3582077 24471. Fax: 358 2077 24350. E-mail: marjaana.hakkinen@evira.fi.

� Published ahead of print on 19 June 2009.

5244

samples of 952 cattle in a survey carried out by the National Veterinary and FoodResearch Institute (currently, the Finnish Food Safety Authority Evira) at 12 of15 Finnish slaughterhouses in 2003 (13). Altogether, 71 of the bovine fecalisolates originated from dairy cattle and 115 from beef cattle. Because most ofthe isolates originated from different farms and because long-term carriage of thesame genotype of C. jejuni in a herd was considered likely, fecal isolates over theentire year were included in the study. Isolates from 262 carcass samples takenonly between May and August 2003 were included, because those isolated duringthe rest of the year could not have been associated with human infections duringthe summer.

Isolates from chickens (n � 43) were obtained from cecal samples taken atslaughter. Two of three Finnish broiler slaughterhouses participated in this study.All 955 process lots slaughtered between May and August 2003 were sampled.One loopful (10 �l) of cecal contents of three to five chickens from each processlot was directly cultured on modified charcoal cefoperazone deoxycholate agar.One isolate from each campylobacter-positive process lot was submitted to Evirafor further investigation.

Identification and genotyping of isolates. The identification of isolates wasbased on standard biochemical tests (19). The human isolates were genotyped atTHL and the bovine and chicken isolates at Evira by PFGE using SmaI forrestriction as described by Hakkinen et al. (13).

All isolates representing overlapping SmaI subtypes were additionally sub-typed using KpnI for restriction. DNA was digested for a minimum of 4 h at 37°Cwith 20 U of KpnI restriction endonuclease (New England Biolabs, Inc., Ipswich,MA) in a final volume of 200 �l containing 2 �l of bovine serum albumin (NewEngland Biolabs, Inc., Ipswich, MA). PFGE data were analyzed with Bionumer-ics V5.10 (Applied Maths, Kortrijk, Belgium) at 0.5% optimization and 1.0%tolerance. Patterns differing by at least a single band were considered differentsubtypes. Subtypes obtained by SmaI and KpnI restriction were named S1, S2,etc., and K1, K2, etc., respectively.

Evaluation of the temporal association among isolates. The temporal associ-ation of the SmaI/KpnI subtypes among isolates from chickens and patients wasevaluated using the criteria presented by Karenlampi et al. (21).

Statistical methods. The �2 test was performed to investigate the associationbetween human C. jejuni genotypes and animal reservoirs as well as their asso-ciation with the type of cattle herds. A P value of �0.05 indicated statisticalsignificance.

RESULTS

We identified 109 different SmaI subtypes among the 419 C.jejuni isolates investigated. Forty-three subtypes were distin-guished among the 175 isolates from human infections, 15subtypes among the 43 isolates from chickens, and 61 subtypesamong the 201 isolates from cattle (data not shown). Of these,26, 10, and 36 occurred only once in human, chicken, andbovine samples, respectively; 18 isolates from humans and 1from chickens were untypeable by SmaI.

Fourteen SmaI subtypes of C. jejuni (32.6% of all 43 humansubtypes) representing 114 (65.1%) of 175 human isolates wereindistinguishable from those of chicken or bovine isolates (Ta-ble 1). In total, 36 (83.7%) of 43 chicken isolates and 62(30.8%) of 201 isolates from cattle represented SmaI subtypesshared with humans.

Further subtyping of 212 C. jejuni isolates (114 human, 36chicken, and 62 cattle isolates), representing the 14 overlap-ping SmaI subtypes, with KpnI as a restriction enzyme yielded44 subtypes, 17 of which were shared between human andanimal isolates (Table 1). The combined type S6/K12 predom-inated among isolates from human patients (12%) and oc-curred in both chickens and cattle (Table 1; Fig. 1).

Of the combined SmaI/KpnI subtypes, 12 were present onlyin humans, 4 only in chickens, and 12 only in cattle. In total, theSmaI/KpnI profiles of 97 (55.4%) human isolates were indis-tinguishable from those of chicken or cattle isolates. The over-lapping combined SmaI/KpnI subtypes accounted for 69.8%

(30/43) and 15.9% (32/201) of the chicken and cattle isolates,respectively. The occurrence of identical SmaI/KpnI subtypeswith human C. jejuni isolates was significantly associated withanimal host species (P � 0.001).

A total of 17 of the 71 (23.9%) fecal isolates from dairycattle and 15 (13.0%) of the 115 fecal isolates from beef cattlerepresented the overlapping SmaI/KpnI subtypes with humanisolates. The occurrence of identical SmaI/KpnI subtypes with

TABLE 1. SmaI/KpnI subtypes of Campylobacter jejuni indomestically acquired sporadic human infections,

chickens, and cattle in Finland betweenJune and August 2003a

PFGE subtype (SmaI/KpnI)No. (%) of isolates from:

Humans Chicken Cattle

S1/K13 0 (0.0) 0 (0.0) 6 (2.9)S1/K21 0 (0.0) 0 (0.0) 1 (0.5)S1/K22 0 (0.0) 0 (0.0) 5 (2.4)S1/K23 0 (0.0) 0 (0.0) 1 (0.5)S1/K24 0 (0.0) 0 (0.0) 4 (1.9)S1/K25 0 (0.0) 0 (0.0) 1 (0.5)S1/K26 0 (0.0) 0 (0.0) 5 (2.4)S1/K33 1 (0.0) 0 (0.0) 0 (0.0)S4/K28 0 (0.0) 1 (2.3) 3 (1.5)S4/K29 1 (0.6) 1 (2.3) 1 (0.5)S4/K31 0 (0.0) 2 (4.7) 0 (0.0)S4/K32 0 (0.0) 1 (2.3) 0 (0.0)S5/K27 1 (0.6) 0 (0.0) 10 (4.9)S6/K12 21 (12.0) 2 (4.7) 7 (3.4)S7/K1 12 (6.9) 2 (4.7) 7 (3.4)S7/K2 4 (2.3) 2 (4.7) 2 (1.0)S7/K3 17 (9.7) 2 (4.7) 1 (0.5)S7/K36 2 (1.1) 0 (0.0) 0 (0.0)S22/K14 0 (0.0) 0 (0.0) 1 (0.5)S22/K15 0 (0.0) 0 (0.0) 1 (0.5)S22/K16 1 (0.6) 0 (0.0) 1 (0.5)S38/K17 0 (0.0) 0 (0.0) 1 (0.5)S38/K34 1 (0.6) 0 (0.0) 0 (0.0)S54/K8 0 (0.0) 0 (0.0) 1 (0.5)S54/K9 0 (0.0) 1 (2.3) 0 (0.0)S54/K10 6 (3.4) 2 (4.7) 0 (0.0)S54/K11 3 (1.7) 1 (2.3) 0 (0.0)S54/K42 1 (0.6) 0 (0.0) 0 (0.0)S54/K43 2 (1.1) 0 (0.0) 0 (0.0)S64/K19 7 (4.0) 1 (2.3) 1 (0.5)S64/K35 2 (1.1) 0 (0.0) 0 (0.0)S66/K18 4 (2.3) 0 (0.0) 1 (0.5)S74/K4 5 (2.9) 8 (18.6) 0 (0.0)S74/K5 8 (4.6) 4 (9.3) 1 (0.5)S74/K6 0 (0.0) 1 (2.3) 0 (0.0)S74/K7 2 (1.1) 2 (4.7) 0 (0.0)S74/K37 1 (0.6) 0 (0.0) 0 (0.0)S74/K38 1 (0.6) 0 (0.0) 0 (0.0)S74/K39 1 (0.6) 0 (0.0) 0 (0.0)S74/K40 1 (0.6) 0 (0.0) 0 (0.0)S76/K20 3 (1.7) 1 (2.3) 0 (0.0)S76/K6 1 (0.6) 0 (0.0) 0 (0.0)S77/K30 1 (0.6) 1 (2.3) 0 (0.0)S77/K41 3 (1.7) 0 (0.0) 0 (0.0)S78/K6 1 (0.6) 1 (2.3) 0 (0.0)

Overlapping combined subtypes 97 (55.4) 30 (69.8) 32 (15.9)

Overlapping SmaI types 114 (65.1) 36 (83.7) 62 (30.8)

Total no. of isolates 175 43 201

a Overlapping subtypes between human and animal isolates appear in bold.

VOL. 75, 2009 CHICKENS AND CATTLE AS CAMPYLOBACTER INFECTION SOURCES 5245

human C. jejuni isolates in cattle was not significantly related toherd type (P � 0.056). All bovine subtypes overlapping thoseof humans occurred among isolates from dairy cattle, with theexception of S22/K16, isolated only from beef cattle (Fig. 2).

A temporal association of the SmaI/KpnI subtypes amongisolates from chickens and patients was possible in 55 (31.4%)of 175 human infections (Table 2). Isolates from 12 (6.9%)human infections temporally associated with chicken isolatesrepresented SmaI/KpnI subtypes that failed to occur in cattle.

DISCUSSION

In this study, we compared the DNA macrorestriction pro-files of C. jejuni isolates from domestic human infections,chickens, and cattle covering the whole of Finland over a timeframe of three summer months with the aim of estimating theattribution of these animal sources to human infections. Atotal of 419 C. jejuni isolates were genotyped with PFGE usingSmaI and KpnI as restriction enzymes.

The C. jejuni isolates from food production animals werecollected from 12 cattle slaughterhouses and 2 chicken slaugh-terhouses, representing 98% of the cattle and 85% of thechicken slaughter volume in Finland in 2003, respectively. The

human clinical C. jejuni isolates of domestic origin represented54% of all isolates collected by 9 of 25 Finnish clinical labo-ratories during a three-month period from June to August2003. The total number of campylobacter infections reportedduring the same time period in Finland was 1,281, includinginfections contracted abroad (http://www3.ktl.fi/stat/).

The summer months were chosen as the time period toexamine because of the pronounced seasonality of humancampylobacteriosis and because the proportion of domesti-cally acquired human cases in Finland is highest during thesummer months (23; www.ktl.fi/attachments/suomi/julkaisut/julkaisusarja_b/2005/2005b13.pdf). Furthermore, the occur-rence of campylobacter in Finnish chicken process lots and,consequently, in retail poultry meat peaks in July and August(7, 15, 24). A comparison of C. jejuni isolates from retailchicken meat would have focused specifically on the genotypesto which consumers are exposed. On the other hand, by sam-pling at slaughter, we could obtain samples from more than

FIG. 1. Distribution of 17 Campylobacter jejuni SmaI/KpnI subtypes among isolates from domestically acquired human infections, chickens, andcattle in Finland between June and August 2003.

FIG. 2. Distribution of the most prevalent SmaI/KpnI subtypes ofhuman Campylobacter jejuni isolates among fecal isolates from Finnishdairy and beef cattle.

TABLE 2. Temporal association between human and broilerCampylobacter jejuni isolates during the seasonal peak in

Finland from June to August 2003

SmaI/KpnI subtype

No. of human isolates temporally associatedwith isolates from positive broiler flocks/total

no. of human isolates

June July August Total

S4/K29 0/0 0/1 0/0 0/1S6/K12 0/0 7/7 14/14 21/21S7/K1 0/1 8/8 3/3 11/12S7/K2 0/0 0/4 0/0 0/4S7/K3 0/2 1/6 9/9 10/17S54/K10 0/0 0/6 0/0 0/6S54/K11 0/0 0/1 1/2 1/3S64/K19 0/0 0/5 1/2 1/7S74/K4 0/0 5/5 0/0 5/5S74/K5 0/0 0/8 0/0 0/8S74/K7 0/0 0/0 2/2 2/2S76/K20 0/0 0/0 3/3 3/3S77/K30 0/0 0/0 1/1 1/1S78/K6 0/0 0/1 0/0 0/1Total 0/3 21/52 34/36 55/91Total no. of human

isolates per month11 106 58 175

5246 HAKKINEN ET AL. APPL. ENVIRON. MICROBIOL.

80% of all process lots during the sampling period and, there-fore, probably a better overall view of the situation. As Nielsenet al. (31) observed, the same C. jejuni subtypes that colonizethe intestines of chickens can be detected in retail samples ofchicken meat. Due to the small number of cecal samples perprocess lot, we may have excluded some positive lots if thecontamination rate of chickens in the process lot was less than50%. In recent years, after the implementation of the Finnishcampylobacter monitoring program for poultry in 2004, slightlyhigher prevalences (5.9 to 7.4%) of campylobacters have beenreported during the summer months, probably due to thehigher number of cecal samples (10 ceca per process lot [7]),than the prevalence in this study, which is in accordance withthat previously reported by Perko-Makela et al. (35). In gen-eral, the prevalence of campylobacters in Finnish chicken pro-cess lots is lower than in most other countries, where preva-lences from 15% to 87% have been reported (7, 29, 38). Theproportion of slaughtering by each slaughterhouse in the pre-ceding year was taken into account in the randomized samplingof cattle (13). The bovine fecal isolates collected throughoutthe entire year were included in our study, as evidence suggeststhat the long-term excretion of the same C. jejuni genotypesoccurs both in dairy herds (14, 26) and in the farm environ-ment (8).

As in several previous studies that have used different geno-typing methods (8, 11, 26, 31), we obtained a wide variety ofdifferent C. jejuni subtypes with PFGE typing using SmaI andKpnI restriction enzymes. All different SmaI subtypes amongmultiple isolates from each bovine sample were included in ourstudy. However, more than one SmaI subtype was present inless than 10% of the samples from cattle (13).

A few SmaI/KpnI subtypes predominated among the humanisolates; the five most frequently detected comprised 37% ofall the human isolates. Two subtypes predominated among thechicken strains, accounting for 27% of the chicken isolates.Isolates representing the most prevalent bovine SmaI subtypes(13), except S1, underwent no further analysis using KpnIrestriction, because no identical SmaI types occurred amongthe human isolates. The predominant SmaI subtype in cattle,S1, was divided into seven KpnI subtypes, indicating that bo-vine isolates may be more evenly distributed among differentsubtypes than those from humans and chickens. This may re-flect the diversity of sources of campylobacters in differentgeographical areas of Finland, where cattle farms are situatedall over the country and chicken production is concentrated inthe western part. Kwan et al. (26) and French et al. (9) havepreviously shown that the transmission of C. jejuni genotypesoccurs over distances of only ca. 1 km at maximum in farmlandarea.

In a study by Karenlampi et al. (22), the degree of overlapwas 61% between human and chicken isolates and 5.7% be-tween human and bovine isolates. Our observation of a higheroverlap between isolates from humans and cattle (15.9%) maybe due to the higher number of bovine isolates in our study butmay also indicate differences in the sources of infection be-tween rural and urban areas. Our isolates were collected fromacross the country, excluding the capital city of Helsinki, andthus covered rural areas more extensively than did the humanisolates analyzed by Karenlampi et al. (22) from the Helsinkidistrict in the southern part of Finland. As Ethelberg et al. (6)

and Garrett et al. (10) have suggested, the relative importanceof poultry as a source of campylobacters may be lower ininfections among the rural population. However, a higher per-centage of chicken isolates (69.8%), compared with that ofbovine strains (15.9%), represented SmaI/KpnI subtypes de-tected in human infections in our study.

SmaI/KpnI subtypes of C. jejuni isolated from chickens andcattle, including shared subtypes, were detected in 52% and42% of human cases, respectively. Gilpin et al. (11) reported asimilar overlap between bovine isolates and human infections.A similar percentage of overlap between campylobacters fromchickens and humans, but much higher (83%) between thosefrom cattle and humans, was observed in a study by Nielsen etal. (31). In our study, subtypes shared by chickens and cattlewere isolated in 40% of the human cases and could haveoriginated from either of the two animal reservoirs or fromsome source common to all three of the hosts. Half of thehuman infections in our study could not be explained by theseanimal reservoirs, which may indicate the existence of addi-tional sources for campylobacteriosis besides chickens and cat-tle, as has been suggested previously (2, 23). On the contrary,based on English data, Wilson et al. (40) estimated that meatproduction animals and poultry are the sources of campy-lobacters in 97% of sporadic infections.

Hopkins et al. (17) concluded that genotypically similar C.jejuni strains are rather able to colonize a range of hosts in-stead of being host specific. Besides the SmaI/KpnI subtypesshared by all three of the hosts in our study, seven C. jejunisubtypes were shared between only humans and poultry andthree between only humans and cattle. These subtypes couldrepresent human pathogenic genotypes adapted to chickensand cattle. On the other hand, numerous subtypes were iden-tified among strains isolated only from cattle and some onlyfrom chickens but not from human infections. This observationreinforces previous suggestions that probably not all C. jejunitypes are pathogenic to humans, but nonpathogenic host-spe-cific types may also exist in animal carriers (8, 9, 17, 23, 27, 34).In addition, the most prevalent of the shared C. jejuni subtypesin cattle, S5/K27, was detected in only one patient. This typecould represent subtypes that are adapted to a specific animalhost and that only occasionally cause disease in humans.

The temporal distribution of isolates from human infectionsand the appearance of indistinguishable SmaI/KpnI subtypesin chicken process lots indicate that up to 31% of the humancases of campylobacteriosis could have been mediated bychickens during the study period. Karenlampi et al. (21) havepresented a similar estimate. C. jejuni isolates from 27 (15.4%)human infections not temporally related to chickens were in-distinguishable from bovine isolates. Taking into account thethree subtypes shared only between humans and cattle (S5/K27, S22/K16, and S66/K18), which occurred in 3.4% of thehuman cases, an estimated 19% of the Finnish human infec-tions could have been caused by C. jejuni strains originatingfrom cattle in the summer of 2003. This estimate should beconsidered with caution, however, because indistinguishablegenotypes may also exist in other animal or environmentalsources not included in this study. In addition, some of thehuman infections temporally associated with chicken isolatescould also have been caused by similar bovine campylobacters.However, this study confirms the conclusion of several authors

VOL. 75, 2009 CHICKENS AND CATTLE AS CAMPYLOBACTER INFECTION SOURCES 5247

from other countries (9, 10, 25, 26, 31, 32) that cattle, inaddition to chickens, can be an important source of C. jejunifor human sporadic infections.

The low-level occurrence of campylobacters in bovine car-casses and beef has been reported in several retail and slaugh-terhouse surveys (13, 31, 38). Therefore, beef is generally notconsidered significant in the transmission of campylobacterio-sis. Our results support this conclusion, as none of the C. jejunistrains isolated from bovine carcasses represented similarSmaI/KpnI subtypes to those of human isolates during thesummer of 2003. Direct contact with cattle, fecally contami-nated drinking and swimming waters, and raw milk have beensuggested as routes of occupational and recreational exposureof rural populations to bovine C. jejuni (6, 10, 11). Drinkingdug-well water and swimming in natural waters have beenidentified as risk factors for domestically acquired humancampylobacteriosis in Finland (37), and significant associationshave been shown between particular sequence-type complexesfrom human infections and contact with cattle as well as theconsumption of unpasteurized milk (23). Most milk is deliveredto dairies (ca. 97% in 2003), and the consumption of unpasteur-ized milk is low in Finland (http://www.matilda.fi/servlet/page?_pageid�501,193&_dad�portal30&_schema�PORTAL30&784_MATILDA_JULKAISUT_4484043.docid�906&784_MATILDA_JULKAISUT_4484043.versio�1170260951). However, oc-casional failures in milking hygiene can lead to the contami-nation of milk by campylobacters and cause family outbreakson dairy cattle farms (36). In Sweden, ruminant density hasproven to be more important than poultry-related factors forhuman campylobacter infections in rural areas (32). The situ-ation may be similar in Finland, where the prevalence ofcampylobacters in chickens is low (7, 35) and cattle are com-mon carriers of campylobacters (13).

Due to our substantial sampling over a limited time frame,we could estimate the relative contribution of two well-knownreservoirs of campylobacters, chickens and cattle, to humancampylobacter infections in Finland during the summer of2003. Although chickens can be considered the most importantsingle source of C. jejuni in sporadic, domestically acquiredinfections, the contribution of cattle appeared notable. Due tooverlapping subtypes among chicken and bovine strains, iso-lates from human infections cannot be directly connected tospecific animal sources through PFGE typing without addi-tional epidemiological investigation. Besides hygienic mea-sures in chicken production, new approaches to restrict theoccurrence of campylobacters in other farm animals may beneeded. However, only about half of the domestic human casescould have originated from the sources examined in our study,and the other half remained unexplained.

ACKNOWLEDGMENTS

This work received partial funding from the Finnish VeterinaryScience Foundation.

We thank Kirsi-Maria Eklund, Maaret Hypponen, Sami-Petteri Kar-jalainen, and Kristiina Kopisto for their technical assistance in subtyp-ing the isolates and Marja-Liisa Hanninen for her comments on themanuscript. We also thank the Finnish broiler industry for their coop-eration.

REFERENCES

1. Brown, P. E., O. F. Christensen, H. E. Clough, P. J. Diggle, C. A. Hart, S.Hazel, R. Kemp, A. J. H. Leatherbarrow, A. Moore, J. Sutherst, J. Turner,

N. J. Williams, E. J. Wright, and N. P. French. 2004. Frequency and spatialdistribution of environmental Campylobacter spp. Appl. Environ. Microbiol.70:6501–6511.

2. Carrique-Mas, J., Y. Andersson, M. Hjertqvist, Å. Svensson, A. Torner, andJ. Giesecke. 2005. Risk factors for domestic sporadic campylobacteriosisamong young children in Sweden. Scand. J. Infect. Dis. 37:101–110.

3. Colles, F. M., K. Jones, R. M. Harding, and M. C. J. Maiden. 2003. Geneticdiversity of Campylobacter jejuni isolates from farm animals and farm envi-ronment. Appl. Environ. Microbiol. 69:7409–7413.

4. Devane, M. L., C. Nicol, A. Ball, J. D. Klena, P. Scholes, J. A. Hudson, M. G.Baker, B. J. Gilpin, N. Garret, and M. G. Savill. 2005. The occurrence ofCampylobacter subtypes in environmental reservoirs and potential transmis-sion routes. J. Appl. Microbiol. 98:980–990.

5. Dingle, K. E., F. M. Colles, R. Ure, J. A. Wagenaar, B. Duim, F. J. Bolton,A. J. Fox, D. R. A. Wareing, and M. C. J. Maiden. 2002. Molecular charac-terization of Campylobacter jejuni clones: a basis of epidemiologic investiga-tion. Emerg. Infect. Dis. 8:949–955.

6. Ethelberg, S., J. Simonsen, P. Gerner-Schmidt, K. E. P. Olsen, and K.Mølbak. 2005. Spatial distribution and registry-based case-control analysis ofCampylobacter infections in Denmark, 1991–2001. Am. J. Epidemiol. 162:1008–1015.

7. European Food Safety Authority. 2007. The community summary report ontrends and sources of zoonoses, zoonotic agents, antimicrobial resistance andfoodborne outbreaks in the European Union in 2006. EFSA J. 130:106–352.

8. Fitzgerald, C., K. Stanley, S. Andrew, and K. Jones. 2001. Use of pulsed-fieldgel electrophoresis and flagellin gene typing in identifying clonal groups ofCampylobacter jejuni and Campylobacter coli in farm and clinical environ-ments. Appl. Environ. Microbiol. 67:1429–1436.

9. French, N., M. Barrigas, P. Brown, P. Ribiero, N. Williams, H. Leatherbar-row, R. Birtles, E. Bolton, P. Fearnhead, and A. Fox. 2005. Spatial epidemi-ology and natural population structure of Campylobacter jejuni colonizing afarmland ecosystem. Environ. Microbiol. 7:1116–1126.

10. Garrett, N., M. L. Devane, J. A. Hudson, C. Nicol, A. Ball, J. D. Klena, P.Scholes, M. G. Baker, B. J. Gilpin, and M. G. Savill. 2007. Statistical com-parison of Campylobacter jejuni subtypes from human cases and environmen-tal sources. J. Appl. Microbiol. 103:2113–2121.

11. Gilpin, B. J., B. Thorrold, P. Scholes, R. D. Longhurst, M. Devane, C. Nicol,S. Walker, B. Robson, and M. Savill. 2008. Comparison of Campylobacterjejuni genotypes from dairy cattle and human sources from the Matamata-Piako District of New Zealand. J. Appl. Microbiol. 105:1354–1360.

12. Gormley, F. J., M. MacRae, K. J. Forbes, I. D. Ogden, J. F. Dallas, and J. C.Strachan. 2008. Has retail chicken played a role in the decline of humancampylobacteriosis? Appl. Environ. Microbiol. 74:383–390.

13. Hakkinen, M., H. Heiska, and M.-L. Hanninen. 2007. Prevalence of Campy-lobacter spp. in cattle in Finland and antimicrobial susceptibility of bovineCampylobacter jejuni strains. Appl. Environ. Microbiol. 73:3232–3238.

14. Hakkinen, M, and M.-L. Hanninen. 26 March 2009. Shedding of Campy-lobacter spp. in Finnish cattle on dairy farms. J. Appl. Microbiol. doi:10.1111/j.1365-2672.2009.04269.

15. Hanninen, M.-L., P. Perko-Makela, A. Pitkala, and H. Rautelin. 2000. Athree-year study of Campylobacter jejuni genotypes in humans with domes-tically acquired infections and in chicken samples from the Helsinki area.J. Clin. Microbiol. 38:1998–2000.

16. Hofshagen, M., and H. Kruse. 2005. Reduction in flock prevalence ofCampylobacter spp. in broilers in Norway after implementation of an actionplan. J. Food Prot. 68:2220–2223.

17. Hopkins, K. L., M. Desai, J. A. Frost, J. Stanley, and J. M. J. Logan. 2004.Fluorescent amplified fragment length polymorphism genotyping of Campy-lobacter jejuni and Campylobacter coli strains and its relationship with hostspecificity, serotyping, and phage typing. J. Clin. Microbiol. 42:229–235.

18. Hopkins, R. S., R. Olmsted, and G. R. Istre. 1984. Endemic Campylobacterjejuni infection in Colorado: identified risk factors. Am. J. Public Health74:249–250.

19. International Organization for Standardization. 2006. International stan-dard ISO 10272-1. Microbiology of food and animal feeding stuffs—hori-zontal method for detection and enumeration of Campylobacter spp. Part 1:detection method. International Organization for Standardization, Geneva,Switzerland.

20. Kapperud, G., E. Skjerve, N. H. Bean, S. M. Ostroff, and J. Lassen. 1992.Risk factors for sporadic Campylobacter infections in southeastern Norway.J. Clin. Microbiol. 30:3117–3121.

21. Karenlampi, R., H. Rautelin, M. Hakkinen, and M.-L. Hanninen. 2003.Temporal and geographical distribution and overlap of Penner heat-stableserotypes and pulsed-field electrophoresis genotypes of Campylobacter jejuniisolates collected from humans and chickens in Finland during a seasonalpeak. J. Clin. Microbiol. 41:4870–4872.

22. Karenlampi, R., H. Rautelin, and M.-L. Hanninen. 2007. Evaluation ofgenetic markers and molecular typing methods for prediction sources ofCampylobacter jejuni and C. coli infections. Appl. Environ. Microbiol. 73:1683–1685.

23. Karenlampi, R., H. Rautelin, D. Schonberg-Norio, L. Paulin, and M.-L.Hanninen. 2007. Longitudinal study of Finnish Campylobacter jejuni and C.

5248 HAKKINEN ET AL. APPL. ENVIRON. MICROBIOL.

coli isolates from humans, using multilocus sequence typing, including com-parison with epidemiological data and isolates from poultry and cattle. Appl.Environ. Microbiol. 73:148–155.

24. Katzav, M., P. Isohanni, M. Lund, M. Hakkinen, and U. Lyhs. 2008. PCRassay for the detection of Campylobacter in marinated and non-marinatedpoultry products. Food Microbiol. 25:908–914.

25. Kwan, P. L. S., M. Barrigas, F. J. Bolton, N. P. French, P. Gowland, R.Kemp, H. Leatherbarrow, M. Upton, and A. J. Fox. 2008. Molecular epide-miology of Campylobacter jejuni populations in dairy cattle, wildlife, and theenvironment in a farmland area. Appl. Environ. Microbiol. 74:5130–5138.

26. Kwan, P. L. S., A. Birtles, F. J. Bolton, N. P. French, S. E. Robinson, L. S.Newbold, M. Upton, and A. J. Fox. 2008. Longitudinal study of the molecularepidemiology of Campylobacter jejuni in cattle on dairy farms. Appl. Environ.Microbiol. 74:3626–3633.

27. Manning, G., C. G. Dowson, M. C. Bagnall, I. H. Ahmed, M. West, and D. G.Newell. 2003. Multilocus sequence typing for comparison of veterinary andhuman isolates of Campylobacter jejuni. Appl. Environ. Microbiol. 69:6370–6379.

28. Meldrum, R. J., J. K. Griffiths, R. M. M. Smith, and M. R. Evans. 2005. Theseasonality of human campylobacter infection and Campylobacter isolatesfrom fresh, retail chicken in Wales. Epidemiol. Infect. 133:49–52.

29. Nadeau, E., S. Messier, and S. Quessy. 2002. Prevalence and comparison ofgenetic profiles of Campylobacter strains isolated from poultry and sporadiccases of campylobacteriosis in humans. J. Food Prot. 65:73–78.

30. National Public Health Institute. 2005. Infectious diseases in Finland, 1995–2004. National Public Health Institute, Helsinki, Finland.

31. Nielsen, E. M., V. Fussing, J. Engberg, and N. L. Nielsen. 2006. MostCampylobacter subtypes from sporadic cases can be found in retail poultryproducts and food animals. Epidemiol. Infect. 134:758–767.

32. Nygård, K., Y. Andersson, J. A. Røttingen, Å. Svensson, J. Lindback, T.Kistemann, and J. Giesecke. 2004. Association between environmental riskfactors and campylobacter infections in Sweden. Epidemiol. Infect. 132:317–325.

33. Nylen, G., F. Dunstan, S. R. Palmer, Y. Andersson, F. Bager, J. Cowden, G.Feierl, Y. Galloway, G. Kapperud, F. Megraud, K. Mølbak, L. R. Petersen,

and P. Ruutu. 2002. The seasonal distribution of campylobacter infection innine European countries and New Zealand. Epidemiol. Infect. 128:383–390.

34. On, S. L. W., E. M. Nielsen, J. Engberg, and M. Madsen. 1998. Validity ofSmaI-defined genotypes of Campylobacter jejuni examined by SalI, KpnI andBamHI polymorphism; evidence of identical clones infecting humans, poul-try, and cattle. Epidemiol. Infect. 120:231–237.

35. Perko-Makela, P., M. Hakkinen, T. Honkanen-Buzalski, and M.-L. Hanni-nen. 2002. Prevalence of campylobacters in chicken flocks during the sum-mer of 1999 in Finland. Epidemiol. Infect. 129:187–192.

36. Schildt, M., S. Savolainen, and M.-L. Hanninen. 2005. Long-lasting Campy-lobacter jejuni contamination of milk associated with gastrointestinal illnessin a farming family. Epidemiol. Infect. 133:1–5.

37. Schonberg-Norio, D., J. Takkinen, M.-L. Hanninen, S.-S. Kaukoranta, L.Mattila, and H. Rautelin. 2004. Swimming and Campylobacter infections.Emerg. Infect. Dis. 10:1474–1477.

38. Stern, N. J., K. L. Hiett, G. A. Alfredsson, K. G. Kristinsson, J. Reiersen, H.Hardardottir, H. Briem, E. Gunnarsson, F. Georgsson, R. Lowman, E.Berndtson, A. M. Lammerding, G. M. Paoli, and M. T. Musgrove. 2003.Campylobacter spp. in Icelandic poultry operations and human disease. Epi-demiol. Infect. 130:23–32.

39. Whyte, P., K. McGill, D. Cowley, R. H. Madden, L. Moran, P. Scates, C.Carroll, A. O’Leary, S. Fanning, J. D. Collins, E. McNamara, J. E. Moore,and M. Cormican. 2004. Occurrence of Campylobacter in retail foods inIreland. Int. J. Food Microbiol. 95:111–118.

40. Wilson, D. J., E. Gabriel, A. H. J. Leatherbarrow, J. Cheesbrough, S. Gee, E.Bolton, A. Fox, P. Fearnhead, C. A. Hart, and P. Diggle. 2008. Tracing thesource of campylobacteriosis. PLoS Genet. 4:1–9.

41. Wingstrand, A., J. Neimann, J. Engberg, E. M. Nielsen, P. Gerner-Schmidt,H. Wegener, and K. Mølbak. 2006. Fresh chicken as main risk factor forcampylobacteriosis, Denmark. Emerg. Infect. Dis. 12:280–284.

42. Zhao, C., B. Ge, J. de Villena, R. Sudler, E. Yeh, S. Zhao, D. White, D.Wagner, and J. Meng. 2001. Prevalence of Campylobacter spp., Escherichiacoli, and Salmonella serovars in retail chicken, turkey, pork, and beef fromthe greater Washington, D.C., area. Appl. Environ. Microbiol. 67:5431–5436.

VOL. 75, 2009 CHICKENS AND CATTLE AS CAMPYLOBACTER INFECTION SOURCES 5249